A decomposition from a substitutable many-to-one matching market to a one-to-one matching marketthanks: We acknowledge the financial support of UNSL through grants 03-2016 and 03-1323, and from Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) through grant PIP 112-200801-00655, and from Agencia Nacional de Promoción Científica y Tecnológica through grant PICT 2017-2355.

Pablo Neme Instituto de Matemática Aplicada San Luis (UNSL-CONICET) and Departamento de Matemática, Universidad Nacional de San Luis, San Luis, Argentina. Emails: [email protected] (P. Neme) and [email protected] (J. Oviedo).    Jorge Oviedo22footnotemark: 2
(November 1, 2024)
Abstract

For a many-to-one market with substitutable preferences on the firm’s side, based on the Aizerman-Malishevski decomposition, we define an associated one-to-one market. Given that the usual notion of stability for a one-to-one market does not fit well for this associated one-to-one market, we introduce a new notion of stability. This notion allows us to establish an isomorphism between the set of stable matchings in the many-to-one market and the matchings in the associated one-to-one market that meet this new stability criterion. Furthermore, we present an adaptation of the well-known deferred acceptance algorithm to compute a matching that satisfies this new notion of stability for the associated one-to-one market.


JEL classification: C78, D47.

Keywords: Many-to-one matchings, Aizermann-Malishevski decomposition, one-to-one matchings, deferred acceptance algorithm, stable set

1 Introduction

Many-to-one markets have been extensively studied in the literature, starting with the college admissions problem and extending to the assignment of medical interns to hospitals, as well as applications in labor markets. A fundamental characteristic of these markets is that institutions, which we will refer to as firms, hold preferences over subsets of agents, whom we will refer to as workers.

In this paper, we introduce a decomposition of a many-to-one market in which firms have substitutable preferences over subsets of workers, transforming it into an associated one-to-one market. We construct this decomposition based on the Aizerman-Malishevski decomposition of substitutable preferences (see Aizerman and Malishevski, 1981). Crawford and Kelso (1982) introduced the notion of substitutability in preferences, which is the weakest condition needed to guarantee the existence of stable matchings. A firm is said to have substitutable preferences if it continues to desire a worker even when other workers become unavailable.

The Aizerman-Malishevski decomposition is well known in the choice literature (e.g., Moulin, 1985) and states that any path-independent choice rule can be represented as the union of choices derived from preference relations over individuals. Chambers and Yenmez (2017) apply the Aizerman-Malishevski decomposition to study path-independent choice rules in a matching context. They use this decomposition to develop a deferred acceptance algorithm for many-to-many matching markets with contracts and to analyze its properties. The Aizerman-Malishevski decomposition models a firm as the union of several “copies” of itself, facilitating the representation of firms’ preferences in substitutable cases. A key feature of this decomposition is its capacity to transform a substitutable preference over a subset of workers into multiple, distinct linear preferences over workers. The fact that copies of the same firm have different linear preferences leads to a situation where the usual notion of stability for one-to-one markets does not apply adequately.111Recall that, in traditional one-to-one matching markets, stability requires individual rationality and the absence of blocking pairs. A matching is individually rational if each agent is assigned to an “acceptable” partner. A matching contains a blocking pair if both agents within that pair mutually prefer each other over their current partners in the matching. In Section 2, we present Example 1, which illustrates a many-to-one market where firms have substitutable preferences, along with its corresponding Aizerman-Malishevski decomposition. In this example, we observe that in the many-to-one market, there are four stable matchings, while in the associated one-to-one market, there is only one stable matching under the usual notion of stability. This discrepancy implies that no relation (e.g., no isomorphism) can be established between the stable matchings in the many-to-one market and those in the associated one-to-one market. For this reason, we adapt the notion of stability and introduce stability* for one-to-one matchings. This new notion resembles the classical notion of stability (i.e., it is individually rational and free of blocking pairs) but is adapted to capture the substitutable preferences inherent in the original market. Building on this new notion, we can establish an isomorphism between the set of stable matchings in the substitutable many-to-one market and the set of stable* matchings in the associated one-to-one market.

To highlight the difference between the usual notion and stability*, we introduce two additional conditions not typically required in one-to-one settings. First, for individual rationality* in this associated market, we impose an envy-free condition among firm-copies: no firm-copy matched to a worker should prefer another worker matched to a different copy of the same firm. Second, in the presence of blocking pairs, we require that the worker in the blocking pair is preferred by the relevant firm-copy over all workers matched to other copies of that same firm.

A common approach to prove the non-emptiness of a stable set is constructive. In the seminal paper by Gale and Shapley (1962), an algorithm is presented —the well-known deferred acceptance algorithm— which constructs a stable matching for traditional one-to-one markets. Although our isomorphism demonstrates that the set of stable* one-to-one matchings is non-empty, we adapt the deferred acceptance algorithm to the related one-to-one market, taking into account the specific characteristics of stability*. Although one-to-one markets are traditionally symmetric, our related one-to-one market is not. This is because it originates from the decomposition of a many-to-one substitutable market. Thus, we present two adapted versions of the deferred acceptance algorithm: one where firm-copies propose and another where workers propose. We show that, whether the firm-copies or the workers are the proposers, the algorithm returns a stable* matching. In this way, independently of the isomorphism, we establish that the set of stable* matchings for the related one-to-one market is non-empty.

The idea of decomposing a many-to-one market into a one-to-one market has been previously studied under a more restrictive preference structure. When firms’ preferences are assumed to be responsive to an individual ranking of workers—a more restrictive structure than substitutable preferences—Roth and Sotomayor (1990) demonstrate that each firm can be decomposed into identical units (or copies) according to its capacity (quotas q𝑞qitalic_q). A key distinction in this decomposition is that each of these copies shares the same individual preferences over workers (derived from responsive preferences), and workers rank all copies of a given firm above those of any other firm, preserving the same order of preferences across all copies. This decomposition transforms a many-to-one market with responsive preferences into a corresponding one-to-one market. Furthermore, Roth and Sotomayor (1990) establish an isomorphism between the stable matchings of a many-to-one market and those of an associated one-to-one market.

The paper is organized as follows. In Section 2, we present the many-to-one market, the Aizerman-Malishevski decomposition, and the associated one-to-one market. For this associated one-to-one market, in Section 3, we present an adapted deferred acceptance algorithm and an isomorphism with the many-to-one market. Finally, in Section 4 some final remarks are presented.

2 Preliminaries

This section contains two subsections. In Subsection 2.1, we present the many-to-one market with substitutable preferences, along with its respective notions of matching, individual rationality, and stability. In Subsection 2.2, we introduce the Aizerman-Malishevski decomposition of substitutable preferences into linear orders, and we show a decomposition of the many-to-one market into an associated one-to-one market with its respective notions of matching, individual rationality*, and stability*.

2.1 Many-to-one matching market

Let 𝓜=(Φ,𝒲,P)𝓜Φ𝒲𝑃\boldsymbol{\mathcal{M}}=(\Phi,\mathcal{W},{P})bold_caligraphic_M = ( roman_Φ , caligraphic_W , italic_P ) be a many-to-one matching market, where 𝚽={φ1,,φn}𝚽subscript𝜑1subscript𝜑𝑛\boldsymbol{\Phi}=\{\varphi_{1},\dots,\varphi_{n}\}bold_Φ = { italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , … , italic_φ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT } denotes the set of n𝑛nitalic_n firms and 𝓦={w1,,wk}𝓦subscript𝑤1subscript𝑤𝑘\boldsymbol{{\mathcal{W}}}=\{w_{1},\dots,w_{k}\}bold_caligraphic_W = { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , … , italic_w start_POSTSUBSCRIPT italic_k end_POSTSUBSCRIPT } the set of k𝑘kitalic_k workers. Each firm φΦ𝜑Φ\varphi\in\Phiitalic_φ ∈ roman_Φ has a strict, transitive, and complete preference relation Pφsubscript𝑃𝜑P_{\varphi}italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT over 2𝒲superscript2𝒲2^{{{\mathcal{W}}}}2 start_POSTSUPERSCRIPT caligraphic_W end_POSTSUPERSCRIPT, and each worker w𝒲𝑤𝒲w\in{{{\mathcal{W}}}}italic_w ∈ caligraphic_W has a strict, transitive, and complete preference relation Pwsubscript𝑃𝑤P_{w}italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT over Φ{}Φ{\Phi}\cup\left\{\emptyset\right\}roman_Φ ∪ { ∅ }. Let Rwsubscript𝑅𝑤R_{w}italic_R start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT denote the weak preference relation associated with Pwsubscript𝑃𝑤P_{w}italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT, meaning that if φiRwφjsubscript𝜑𝑖subscript𝑅𝑤subscript𝜑𝑗\varphi_{i}R_{w}\varphi_{j}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_R start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT imply that either φiPwφjsubscript𝜑𝑖subscript𝑃𝑤subscript𝜑𝑗\varphi_{i}P_{w}\varphi_{j}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT or φi=φjsubscript𝜑𝑖subscript𝜑𝑗\varphi_{i}=\varphi_{j}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT = italic_φ start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT. The collection of these preference relations forms the preference profile 𝑷=((Pφ)φΦ,(Pw)w𝒲)𝑷subscriptsubscript𝑃𝜑𝜑Φsubscriptsubscript𝑃𝑤𝑤𝒲\boldsymbol{P}=\left((P_{\varphi})_{\varphi\in\Phi},(P_{w})_{w\in{{\mathcal{W}% }}}\right)bold_italic_P = ( ( italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ) start_POSTSUBSCRIPT italic_φ ∈ roman_Φ end_POSTSUBSCRIPT , ( italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ) start_POSTSUBSCRIPT italic_w ∈ caligraphic_W end_POSTSUBSCRIPT ). Given a subset of workers W𝒲𝑊𝒲W\subseteq{{{\mathcal{W}}}}italic_W ⊆ caligraphic_W, let Cφ(W)subscript𝐶𝜑𝑊C_{\varphi}(W)italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ) denote firm φ𝜑\varphiitalic_φ’s choice function induced by preference relation Pφsubscript𝑃𝜑P_{\varphi}italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT; that is, Cφ(W)={W𝒲:W=maxW′′WPφ}subscript𝐶𝜑𝑊conditional-setsuperscript𝑊𝒲superscript𝑊subscriptsuperscript𝑊′′𝑊subscript𝑃𝜑C_{\varphi}(W)=\{W^{\prime}\subseteq{{\mathcal{W}}}:W^{\prime}=\max_{W^{\prime% \prime}\subseteq W}P_{\varphi}\}italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ) = { italic_W start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ⊆ caligraphic_W : italic_W start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = roman_max start_POSTSUBSCRIPT italic_W start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT ⊆ italic_W end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT }. Firm φ𝜑\varphiitalic_φ’s preferences are substitutable if for each W𝒲𝑊𝒲W\subseteq\mathcal{W}italic_W ⊆ caligraphic_W we have that wCφ(W)𝑤subscript𝐶𝜑𝑊w\in C_{\varphi}(W)italic_w ∈ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ) implies that wCφ(W{w})𝑤subscript𝐶𝜑𝑊superscript𝑤w\in C_{\varphi}(W\setminus\{w^{\prime}\})italic_w ∈ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ∖ { italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT } ) for each wW{w}superscript𝑤𝑊𝑤w^{\prime}\in W\setminus\{w\}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_W ∖ { italic_w }. Note that, if preference Pφsubscript𝑃𝜑P_{\varphi}italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT is substitutable, then the induced choice function Cφ()subscript𝐶𝜑C_{\varphi}(\cdot)italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( ⋅ ) is not only substitutable but also satisfies consistency. We say that a choice function Cφ()subscript𝐶𝜑C_{\varphi}(\cdot)italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( ⋅ ) satisfies consistency if Cφ(X)YXCφ(X)=Cφ(Y)subscript𝐶𝜑𝑋𝑌𝑋subscript𝐶𝜑𝑋subscript𝐶𝜑𝑌C_{\varphi}(X)\subseteq Y\subseteq X\implies C_{\varphi}(X)=C_{\varphi}(Y)italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_X ) ⊆ italic_Y ⊆ italic_X ⟹ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_X ) = italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_Y ). Moreover, if the choice function Cφ()subscript𝐶𝜑C_{\varphi}(\cdot)italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( ⋅ ) satisfies substitutability and consistency, then it also satisfies path-independence (see Alkan, 2002). Path-independence property states that given a firm φ𝜑\varphiitalic_φ and two subsets W𝑊Witalic_W and Wsuperscript𝑊W^{\prime}italic_W start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT of workers,

Cφ(WW)=Cφ(Cφ(W)W).subscript𝐶𝜑𝑊superscript𝑊subscript𝐶𝜑subscript𝐶𝜑𝑊superscript𝑊C_{\varphi}(W\cup W^{\prime})=C_{\varphi}(C_{\varphi}(W)\cup W^{\prime}).italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ∪ italic_W start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ) = italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_W ) ∪ italic_W start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ) . (1)

Throughout the paper, we assume that firms in the many-to-one matching market have substitutable preferences.

A many-to-one matching is a mapping μ:Φ𝒲2Φ𝒲:𝜇Φ𝒲superscript2Φ𝒲\mu:\Phi\cup{{\mathcal{W}}}\rightarrow 2^{\Phi\cup{{\mathcal{W}}}}italic_μ : roman_Φ ∪ caligraphic_W → 2 start_POSTSUPERSCRIPT roman_Φ ∪ caligraphic_W end_POSTSUPERSCRIPT that satisfies the following conditions: (i) μ(w)Φ𝜇𝑤Φ\mu(w)\in\Phi\cup{\emptyset}italic_μ ( italic_w ) ∈ roman_Φ ∪ ∅ for each w𝒲𝑤𝒲w\in{{\mathcal{W}}}italic_w ∈ caligraphic_W, (ii) μ(φ)2𝒲𝜇𝜑superscript2𝒲\mu(\varphi)\in 2^{{{\mathcal{W}}}}italic_μ ( italic_φ ) ∈ 2 start_POSTSUPERSCRIPT caligraphic_W end_POSTSUPERSCRIPT for each φΦ𝜑Φ\varphi\in\Phiitalic_φ ∈ roman_Φ, and (iii) μ(w)=φ𝜇𝑤𝜑\mu(w)=\varphiitalic_μ ( italic_w ) = italic_φ if and only if wμ(φ)𝑤𝜇𝜑w\in\mu(\varphi)italic_w ∈ italic_μ ( italic_φ ). A matching μ𝜇\muitalic_μ is blocked by a worker w𝑤witalic_w if Pwμ(w)subscript𝑃𝑤𝜇𝑤\emptyset P_{w}\mu(w)∅ italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_μ ( italic_w ). Similarly, μ𝜇\muitalic_μ is blocked by a firm φ𝜑\varphiitalic_φ if μ(φ)Cφ(μ(φ))𝜇𝜑subscript𝐶𝜑𝜇𝜑\mu(\varphi)\neq C_{\varphi}(\mu(\varphi))italic_μ ( italic_φ ) ≠ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_μ ( italic_φ ) ). We say that a matching is individually rational if it is not blocked by any individual agent. A matching μ𝜇\muitalic_μ is blocked by a worker-firm pair (w,φ)𝑤𝜑(w,\varphi)( italic_w , italic_φ ) if wμ(φ)𝑤𝜇𝜑w\notin\mu(\varphi)italic_w ∉ italic_μ ( italic_φ ), wCφ(μ(φ){w})𝑤subscript𝐶𝜑𝜇𝜑𝑤w\in C_{\varphi}(\mu(\varphi)\cup\{w\})italic_w ∈ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_μ ( italic_φ ) ∪ { italic_w } ), and φPwμ(w)𝜑subscript𝑃𝑤𝜇𝑤\varphi P_{w}\mu(w)italic_φ italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_μ ( italic_w ). Finally, we say that a matching μ𝜇\muitalic_μ is stable if it is not blocked by any individual agent or by any worker-firm pair. We denote by 𝑺(𝓜)𝑺𝓜\boldsymbol{S({{\mathcal{M}}})}bold_italic_S bold_( bold_caligraphic_M bold_) to the set of all stable matchings for the matching market {{\mathcal{M}}}caligraphic_M.

2.2 An associated one-to-one market

In this subsection, we present the Aizerman-Malishevski decomposition (see Chambers and Yenmez, 2017). This decomposition allows us to establish a connection between a many-to-one market with substitutable preferences and a one-to-one market. It captures the essence of the substitutability condition on firms’ preferences by splitting each firm into multiple copies and making that each copy of a firm behaves independently of the others, meaning that the copies have different linear preferences over workers.

Aizermann-Malishevski decomposition:

If Pφsubscript𝑃𝜑P_{\varphi}italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT is a substitutable preference relation on 2𝒲superscript2𝒲2^{{\mathcal{W}}}2 start_POSTSUPERSCRIPT caligraphic_W end_POSTSUPERSCRIPT, then there is a finite sequence of preference relations {Pφj}jJφsubscriptsubscript𝑃𝜑𝑗𝑗subscript𝐽𝜑\{P_{\varphi j}\}_{j\in J_{\varphi}}{ italic_P start_POSTSUBSCRIPT italic_φ italic_j end_POSTSUBSCRIPT } start_POSTSUBSCRIPT italic_j ∈ italic_J start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT end_POSTSUBSCRIPT on 𝒲𝒲{{\mathcal{W}}}caligraphic_W such that, for all S𝒲𝑆𝒲S\subseteq{{\mathcal{W}}}italic_S ⊆ caligraphic_W:

Cφ(S)=jJφmaxSPφjsubscript𝐶𝜑𝑆subscript𝑗subscript𝐽𝜑subscript𝑆subscript𝑃𝜑𝑗C_{\varphi}(S)=\bigcup_{j\in J_{\varphi}}\max_{S}P_{\varphi j}italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_S ) = ⋃ start_POSTSUBSCRIPT italic_j ∈ italic_J start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT end_POSTSUBSCRIPT roman_max start_POSTSUBSCRIPT italic_S end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_φ italic_j end_POSTSUBSCRIPT

Next, using the Aizerman-Malishevski decomposition, we introduce a related one-to-one market, denoted by 𝑴𝑴\boldsymbol{M}bold_italic_M, and provide a notion of stability suitable for this newly associated market. This decomposition allows us to establish a link with the many-to-one market with substitutable preferences. Let

={f11,,f1J1,,fn1,,fnJn}subscript𝑓11subscript𝑓1subscript𝐽1subscript𝑓𝑛1subscript𝑓𝑛subscript𝐽𝑛{{\mathcal{F}}}=\{f_{11},\dots,f_{1J_{1}},\dots,f_{n1},\dots,f_{nJ_{n}}\}caligraphic_F = { italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT , … , italic_f start_POSTSUBSCRIPT 1 italic_J start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT , … , italic_f start_POSTSUBSCRIPT italic_n 1 end_POSTSUBSCRIPT , … , italic_f start_POSTSUBSCRIPT italic_n italic_J start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT end_POSTSUBSCRIPT }

be the set of firm-copies, where fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT denotes the firm φisubscript𝜑𝑖\varphi_{i}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT’s j𝑗jitalic_j-th copy. These copies are taken from the Aizermann-Malishevski decomposition, so that Cφi(S)=jJimaxSPijsubscript𝐶subscript𝜑𝑖𝑆subscript𝑗subscript𝐽𝑖subscript𝑆subscript𝑃𝑖𝑗C_{\varphi_{i}}(S)=\bigcup_{j\in J_{i}}\max_{S}P_{ij}italic_C start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( italic_S ) = ⋃ start_POSTSUBSCRIPT italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT roman_max start_POSTSUBSCRIPT italic_S end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT.222Henceforth, to easy notation, we write Pijsubscript𝑃𝑖𝑗P_{ij}italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT instead of Pφij.subscript𝑃subscript𝜑𝑖𝑗P_{\varphi_{i}j}.italic_P start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT . The set of workers is 𝒲={w1,,wk}𝒲subscript𝑤1subscript𝑤𝑘{{\mathcal{W}}}=\{w_{1},\dots,w_{k}\}caligraphic_W = { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , … , italic_w start_POSTSUBSCRIPT italic_k end_POSTSUBSCRIPT }. Now, each worker w𝑤witalic_w has a strict, transitive, and complete preference relation P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT on the set {}{{\mathcal{F}}}\cup\{\emptyset\}caligraphic_F ∪ { ∅ } such that:

  1. (1)

    φiPwφisubscript𝜑𝑖subscript𝑃𝑤subscript𝜑superscript𝑖\varphi_{i}P_{w}\varphi_{i^{\prime}}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT implies that fijP¯wfijsubscript𝑓𝑖𝑗subscript¯𝑃𝑤subscript𝑓superscript𝑖superscript𝑗f_{ij}\overline{P}_{w}f_{i^{\prime}j^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT for each jJi𝑗subscript𝐽𝑖j\in J_{i}italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT and each jJisuperscript𝑗subscript𝐽superscript𝑖j^{\prime}\in J_{i^{\prime}}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT.

  2. (2)

    fijP¯wfij if and only if j<jsubscript𝑓𝑖𝑗subscript¯𝑃𝑤subscript𝑓𝑖superscript𝑗 if and only if 𝑗superscript𝑗f_{ij}\overline{P}_{w}f_{ij^{\prime}}\text{ if and only if }j<j^{\prime}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT if and only if italic_j < italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT, for each iI={1,,n}𝑖𝐼1𝑛i\in I=\{1,\dots,n\}italic_i ∈ italic_I = { 1 , … , italic_n } and each j,jJi.𝑗superscript𝑗subscript𝐽𝑖j,j^{\prime}\in J_{i}.italic_j , italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT .

Condition (1) ensures that workers’ preferences (P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT) over firm-copies reflect exactly the preference relation (Pwsubscript𝑃𝑤P_{w}italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT) over firms. In turn, condition (2) ensures that P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT is a linear order. We denote the associated one-to-one matching market as M=(,𝒲,(P¯w)wW,(Pij)iI,jJi)𝑀𝒲subscriptsubscript¯𝑃𝑤𝑤𝑊subscriptsubscript𝑃𝑖𝑗formulae-sequence𝑖𝐼𝑗subscript𝐽𝑖M=({{\mathcal{F}}},{{\mathcal{W}}},(\overline{P}_{w})_{w\in W},(P_{ij})_{i\in I% ,j\in J_{i}})italic_M = ( caligraphic_F , caligraphic_W , ( over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ) start_POSTSUBSCRIPT italic_w ∈ italic_W end_POSTSUBSCRIPT , ( italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) start_POSTSUBSCRIPT italic_i ∈ italic_I , italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT ).

The following example illustrates how we can decompose a many-to-one market with substitutable preferences for firms into an associated one-to-one market using the Aizermann-Malishevski decomposition.

Example 1

Consider a many-to-one matching market (Φ,𝒲,P)Φ𝒲𝑃(\Phi,\mathcal{W},P)( roman_Φ , caligraphic_W , italic_P ), where Φ={φ1,φ2}Φsubscript𝜑1subscript𝜑2\Phi=\{\varphi_{1},\varphi_{2}\}roman_Φ = { italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT }, 𝒲={w1,w2,w3,w4}𝒲subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\mathcal{W}=\{w_{1},w_{2},w_{3},w_{4}\}caligraphic_W = { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT }, and the preference profile is presented in the following table:333To ease notation, in the firms’ preference over 2𝒲superscript2𝒲2^{\mathcal{W}}2 start_POSTSUPERSCRIPT caligraphic_W end_POSTSUPERSCRIPT, we enlist only the acceptable subsets and we omit the curly brackets, i.e. we write Pφ:w1w2,w1,w2:subscript𝑃𝜑subscript𝑤1subscript𝑤2subscript𝑤1subscript𝑤2P_{\varphi}:w_{1}w_{2},w_{1},w_{2}italic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT instead of writing Pφ:{w1w2},{w1},{w2},:subscript𝑃𝜑subscript𝑤1subscript𝑤2subscript𝑤1subscript𝑤2P_{\varphi}:\{w_{1}w_{2}\},\{w_{1}\},\{w_{2}\},\emptysetitalic_P start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT : { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT } , { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT } , { italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT } , ∅. For the workers’ preferences, we also enlist only acceptable firms.

Pφ1:w1w2,w1w3,w2w4,w3w4,w1,w2,w3,w4:subscript𝑃subscript𝜑1subscript𝑤1subscript𝑤2subscript𝑤1subscript𝑤3subscript𝑤2subscript𝑤4subscript𝑤3subscript𝑤4subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\displaystyle P_{\varphi_{1}}:w_{1}w_{2},w_{1}w_{3},w_{2}w_{4},w_{3}w_{4},w_{1% },w_{2},w_{3},w_{4}italic_P start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT Pw1=Pw2:φ2,φ1:subscript𝑃subscript𝑤1subscript𝑃subscript𝑤2subscript𝜑2subscript𝜑1\displaystyle P_{w_{1}}=P_{w_{2}}:\varphi_{2},\varphi_{1}italic_P start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT = italic_P start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT
Pφ2:w3w4,w1w3,w2w4,w1w2,w4,w3,w2,w1:subscript𝑃subscript𝜑2subscript𝑤3subscript𝑤4subscript𝑤1subscript𝑤3subscript𝑤2subscript𝑤4subscript𝑤1subscript𝑤2subscript𝑤4subscript𝑤3subscript𝑤2subscript𝑤1\displaystyle P_{\varphi_{2}}:w_{3}w_{4},w_{1}w_{3},w_{2}w_{4},w_{1}w_{2},w_{4% },w_{3},w_{2},w_{1}italic_P start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT Pw3=Pw4:φ1,φ2:subscript𝑃subscript𝑤3subscript𝑃subscript𝑤4subscript𝜑1subscript𝜑2\displaystyle P_{w_{3}}=P_{w_{4}}:\varphi_{1},\varphi_{2}italic_P start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_POSTSUBSCRIPT = italic_P start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT

The associated Aizermann-Malishevski decomposition of the substitutable preferences into linear preferences is presented in the following table:

P11:w1,w2,w3,w4:subscript𝑃11subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\displaystyle P_{11}:w_{1},w_{2},w_{3},w_{4}italic_P start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT P21:w3,w4,w2,w1:subscript𝑃21subscript𝑤3subscript𝑤4subscript𝑤2subscript𝑤1\displaystyle P_{21}:w_{3},w_{4},w_{2},w_{1}italic_P start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT
P12:w1,w2,w4,w3:subscript𝑃12subscript𝑤1subscript𝑤2subscript𝑤4subscript𝑤3\displaystyle P_{12}:w_{1},w_{2},w_{4},w_{3}italic_P start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT P22:w3,w4,w1,w2:subscript𝑃22subscript𝑤3subscript𝑤4subscript𝑤1subscript𝑤2\displaystyle P_{22}:w_{3},w_{4},w_{1},w_{2}italic_P start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT
P13:w1,w4,w2,w3:subscript𝑃13subscript𝑤1subscript𝑤4subscript𝑤2subscript𝑤3\displaystyle P_{13}:w_{1},w_{4},w_{2},w_{3}italic_P start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT P23:w3,w2,w4,w1:subscript𝑃23subscript𝑤3subscript𝑤2subscript𝑤4subscript𝑤1\displaystyle P_{23}:w_{3},w_{2},w_{4},w_{1}italic_P start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT
P14:w2,w1,w3,w4:subscript𝑃14subscript𝑤2subscript𝑤1subscript𝑤3subscript𝑤4\displaystyle P_{14}:w_{2},w_{1},w_{3},w_{4}italic_P start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT P24:w4,w3,w2,w1:subscript𝑃24subscript𝑤4subscript𝑤3subscript𝑤2subscript𝑤1\displaystyle P_{24}:w_{4},w_{3},w_{2},w_{1}italic_P start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT
P15:w2,w1,w4,w3:subscript𝑃15subscript𝑤2subscript𝑤1subscript𝑤4subscript𝑤3\displaystyle P_{15}:w_{2},w_{1},w_{4},w_{3}italic_P start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT P25:w4,w3,w1,w2:subscript𝑃25subscript𝑤4subscript𝑤3subscript𝑤1subscript𝑤2\displaystyle P_{25}:w_{4},w_{3},w_{1},w_{2}italic_P start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT
P16:w2,w3,w1,w4:subscript𝑃16subscript𝑤2subscript𝑤3subscript𝑤1subscript𝑤4\displaystyle P_{16}:w_{2},w_{3},w_{1},w_{4}italic_P start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT P26:w4,w1,w3,w2:subscript𝑃26subscript𝑤4subscript𝑤1subscript𝑤3subscript𝑤2\displaystyle P_{26}:w_{4},w_{1},w_{3},w_{2}italic_P start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT : italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT

Now, the preferences of workers over firm-copies are presented in the following table:

P¯w1=P¯w2:f21,f22,f23,f24,f25,f26,f11,f12,f13,f14,f15,f16:subscript¯𝑃subscript𝑤1subscript¯𝑃subscript𝑤2subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16\displaystyle\overline{P}_{w_{1}}=\overline{P}_{w_{2}}:f_{21},f_{22},f_{23},f_% {24},f_{25},f_{26},f_{11},f_{12},f_{13},f_{14},f_{15},f_{16}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT = over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT
P¯w3=P¯w4:f11,f12,f13,f14,f15,f16,f21,f22,f23,f24,f25,f26:subscript¯𝑃subscript𝑤3subscript¯𝑃subscript𝑤4subscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26\displaystyle\overline{P}_{w_{3}}=\overline{P}_{w_{4}}:f_{11},f_{12},f_{13},f_% {14},f_{15},f_{16},f_{21},f_{22},f_{23},f_{24},f_{25},f_{26}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_POSTSUBSCRIPT = over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_POSTSUBSCRIPT : italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT

Thus, the one-to-one associated market is denoted by (,𝒲,P,P¯𝒲),𝒲subscript𝑃subscript¯𝑃𝒲(\mathcal{F},\mathcal{W},P_{\mathcal{F}},\overline{P}_{\mathcal{W}}),( caligraphic_F , caligraphic_W , italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT , over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ) , where

={f11,,f16,f21,,f26}subscript𝑓11subscript𝑓16subscript𝑓21subscript𝑓26\mathcal{F}=\{f_{11},\ldots,f_{16},f_{21},\ldots,f_{26}\}caligraphic_F = { italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT , … , italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT , … , italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT }

and Psubscript𝑃P_{\mathcal{F}}italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT is the preference profile derived of the Aizermann-Malishevski decomposition. \Diamond

A matching in the associated one-to-one matching market is a mapping λ:𝒲𝒲{}:𝜆𝒲𝒲\lambda:{{\mathcal{F}}}\cup{{\mathcal{W}}}\rightarrow{{\mathcal{F}}}\cup{{% \mathcal{W}}}\cup\{\emptyset\}italic_λ : caligraphic_F ∪ caligraphic_W → caligraphic_F ∪ caligraphic_W ∪ { ∅ } such that λ(f)𝒲{}𝜆𝑓𝒲\lambda(f)\in{{\mathcal{W}}}\cup\{\emptyset\}italic_λ ( italic_f ) ∈ caligraphic_W ∪ { ∅ }, λ(w){}𝜆𝑤\lambda(w)\in{{\mathcal{F}}}\cup\{\emptyset\}italic_λ ( italic_w ) ∈ caligraphic_F ∪ { ∅ } and λ(f)=wλ(w)=fiff𝜆𝑓𝑤𝜆𝑤𝑓\lambda(f)=w\iff\lambda(w)=fitalic_λ ( italic_f ) = italic_w ⇔ italic_λ ( italic_w ) = italic_f.

Given that one of our goals in this paper is to show an isomorphism between the set of stable matchings in the many-to-one substitutable market and a one-to-one market, it is necessary to adapt the notions of individual rationality and stability for the associated one-to-one market. The following example illustrates this need:

Example 1 (continued) The stable matchings of the many-to-one substitutable market are:

μΦ=(φ1φ2w1w2w3w4)μ1=(φ1φ2w1w3w2w4)subscript𝜇Φmatrixsubscript𝜑1subscript𝜑2subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4subscript𝜇1matrixsubscript𝜑1subscript𝜑2subscript𝑤1subscript𝑤3subscript𝑤2subscript𝑤4\displaystyle\mu_{\Phi}=\begin{pmatrix}\varphi_{1}&\varphi_{2}\\ w_{1}w_{2}&w_{3}w_{4}\end{pmatrix}\mu_{1}=\begin{pmatrix}\varphi_{1}&\varphi_{% 2}\\ w_{1}w_{3}&w_{2}w_{4}\end{pmatrix}italic_μ start_POSTSUBSCRIPT roman_Φ end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG ) italic_μ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG )
μ2=(φ1φ2w2w4w1w3)μ𝒲=(φ1φ2w3w4w1w2)subscript𝜇2matrixsubscript𝜑1subscript𝜑2subscript𝑤2subscript𝑤4subscript𝑤1subscript𝑤3subscript𝜇𝒲matrixsubscript𝜑1subscript𝜑2subscript𝑤3subscript𝑤4subscript𝑤1subscript𝑤2\displaystyle\mu_{2}=\begin{pmatrix}\varphi_{1}&\varphi_{2}\\ w_{2}w_{4}&w_{1}w_{3}\end{pmatrix}\mu_{\mathcal{W}}=\begin{pmatrix}\varphi_{1}% &\varphi_{2}\\ w_{3}w_{4}&w_{1}w_{2}\end{pmatrix}italic_μ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG ) italic_μ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG )

Let λ𝜆\lambdaitalic_λ be a matching in the related one-to-one market. Consider the standard notion of individual rationality, i.e., λ(fij)Rij𝜆subscript𝑓𝑖𝑗subscript𝑅𝑖𝑗\lambda(f_{ij})R_{ij}\emptysetitalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_R start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅ and λ(w)R¯w𝜆𝑤subscript¯𝑅𝑤\lambda(w)\overline{R}_{w}\emptysetitalic_λ ( italic_w ) over¯ start_ARG italic_R end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∅, and the standard notion of stability, i.e., there is no blocking pair (fij,w)subscript𝑓𝑖𝑗𝑤(f_{ij},w)( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT , italic_w ) such that wλ(fij)𝑤𝜆subscript𝑓𝑖𝑗w\neq\lambda(f_{ij})italic_w ≠ italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), wPijλ(fij)𝑤subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗wP_{ij}\lambda(f_{ij})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), and fijP¯wλ(w)subscript𝑓𝑖𝑗subscript¯𝑃𝑤𝜆𝑤f_{ij}\overline{P}_{w}\lambda(w)italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_λ ( italic_w ).

We can observe that the only one-to-one matching in the related market that satisfies this notion of individual rationality and that has no blocking pairs with respect to this blocking notion is the following:

λ=(f11f12f13f14f15f16f21f22f23f24f25f26w3w4w2w1).𝜆matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤3subscript𝑤4subscript𝑤2subscript𝑤1\lambda=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{12}&f_{13}&f_{1% 4}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{3}&w_{4}&\emptyset&\emptyset&\emptyset&\emptyset&w_{2}&w_{1}&\emptyset&% \emptyset&\emptyset&\emptyset\end{pmatrix}.italic_λ = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) .

Note that if we consider any one-to-one matching that assigns workers w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT with any firm-copy f1,subscript𝑓1f_{1,\cdot}italic_f start_POSTSUBSCRIPT 1 , ⋅ end_POSTSUBSCRIPT, and workers w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT and w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT with any firm-copy f2,subscript𝑓2f_{2,\cdot}italic_f start_POSTSUBSCRIPT 2 , ⋅ end_POSTSUBSCRIPT, it will have blocking pairs: any unmatched firm-copy f1subscript𝑓1f_{1\cdot}italic_f start_POSTSUBSCRIPT 1 ⋅ end_POSTSUBSCRIPT with workers w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT or w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT, and similarly, any unmatched firm-copy f2subscript𝑓2f_{2\cdot}italic_f start_POSTSUBSCRIPT 2 ⋅ end_POSTSUBSCRIPT with workers w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT or w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT. This situation clearly implies that, with these notions of individual rationality and stability, it is not possible to define an isomorphism between a set with four elements (the set of stable matchings in the many-to-one substitutable market) and a set with only one element (the set of stable matchings under these notions). \Diamond

To conclude this section, we present the necessary definitions to establish a proper notion of “individual rationality” and “stability”, in order to show in the following section that the sets of “stable matchings” in both markets are isomorphic.

Definition 1

A matching λ𝜆\lambdaitalic_λ is blocked* by a worker wwwitalic_w if \emptyset P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT λ(w)𝜆𝑤\lambda\left(w\right)italic_λ ( italic_w ). On the other hand, a matching λ𝜆\lambdaitalic_λ is blocked* by a firm-copy fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT if either

  1. (1)

    Pijλ(fij)subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗\emptyset P_{ij}\lambda(f_{ij})∅ italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), or

  2. (2)

    λ(fij)𝜆subscript𝑓𝑖𝑗\lambda(f_{ij})\neq\emptysetitalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) ≠ ∅ and there is jJisuperscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT such that λ(fij)Pijλ(fij).𝜆subscript𝑓𝑖superscript𝑗subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗\lambda(f_{ij^{\prime}})P_{ij}\lambda(f_{ij}).italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) .

Note that the absence of blocking* firm-copies implies the negation of condition (2). So, the negation of this condition can be interpreted as an envy-free-like property among the matched copies of a given firm, i.e., no copy that is matched wants to switch partners with another copy. Thus, a one-to-one matching that is not blocked* by any agent we say that it is Individually rational*.

Definition 2

A matching λ𝜆\lambdaitalic_λ is said to be blocked* by a firm-copy and worker pair (fij,w)subscript𝑓𝑖𝑗𝑤(f_{ij},w)( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT , italic_w ) if

  1. 1.

    fijP¯wλ(w)subscript𝑓𝑖𝑗subscript¯𝑃𝑤𝜆𝑤f_{ij}\overline{P}_{w}\lambda(w)italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_λ ( italic_w ); and

  2. 2.

    wPijλ(fij)𝑤subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖superscript𝑗wP_{ij}\lambda(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJi.superscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}.italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT .

Now, we are in a position to formally define the notion of stability* for the associated one-to-one market.

Definition 3

A matching λ𝜆\lambdaitalic_λ is stable* if it is not blocked by any individual agent or any (firm-copy)-worker pair.

Let 𝑺(𝑴)superscript𝑺𝑴\boldsymbol{S^{*}(M)}bold_italic_S start_POSTSUPERSCRIPT bold_∗ end_POSTSUPERSCRIPT bold_( bold_italic_M bold_) denote the set of stable* matchings of the associated one-to-one matching market.

Note that it is important to highlight that condition (2) in Definition 2 is stronger than the usual notion of pairwise block, since it has to account for the fact that some other copy of the same firm may be matched to a better worker. This is particularly relevant in establishing the equivalence (in an isomorphic sense) between the set of stable matchings in the many-to-one matching market and the set of stable* matchings in the associated one-to-one market.

In the following remark, we summarize the possible scenarios in which a matching in the associated one-to-one market is not stable*.

Remark 1

If λS(M)𝜆superscript𝑆𝑀\lambda\notin S^{*}\left(M\right)italic_λ ∉ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ), then (at least) one of the following must hold:

  1. (i)

    There is w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W such that P¯wλ(w)subscript¯𝑃𝑤𝜆𝑤\emptyset\overline{P}_{w}\lambda\left(w\right)∅ over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_λ ( italic_w ).

  2. (ii)

    There is fijsubscript𝑓𝑖𝑗f_{ij}\in\mathcal{F}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F such that Pijλ(fij)subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗\emptyset P_{ij}\lambda\left(f_{ij}\right)∅ italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ).

  3. (iii)

    There are fij,fijsubscript𝑓𝑖𝑗subscript𝑓𝑖superscript𝑗f_{ij},f_{ij^{\prime}}\in\mathcal{F}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ∈ caligraphic_F such that λ(fij)Pijλ(fij)Pij𝜆subscript𝑓𝑖superscript𝑗subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗\lambda(f_{ij^{\prime}})P_{ij}\lambda(f_{ij})P_{ij}\emptysetitalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅.

  4. (iv)

    There is a pair (fij,w)×𝒲subscript𝑓𝑖𝑗𝑤𝒲(f_{ij},w)\in\mathcal{F}\times\mathcal{W}( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT , italic_w ) ∈ caligraphic_F × caligraphic_W such that fijP¯wλ(w)subscript𝑓𝑖𝑗subscript¯𝑃𝑤𝜆𝑤f_{ij}\overline{P}_{w}\lambda(w)italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_λ ( italic_w ) and wPijλ(fij)𝑤subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖superscript𝑗wP_{ij}\lambda(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jsuperscript𝑗j^{\prime}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT.

3 Results on the set of stable* matchings

In this section, we present the main results of the paper. In Subsection 3.1, we adapt the well-known deferred acceptance algorithm, initially introduced in the seminal work of Gale and Shapley (1962), to our one-to-one market framework and the notion of stable*. We show that in both cases, whether firms-copies propose or workers propose, the output of the adapted deferred acceptance algorithm is a stable* matching. Thus, we demonstrate that the set of stable* matchings is non-empty.

In Subsection 3.2, we establish an isomorphism between the set of stable matchings in a many-to-one market with substitutable preferences and the set of stable* matchings in the associated one-to-one market.

3.1 Adapted deferred acceptance algorithm

In this subsection, we introduce two variants of the deferred acceptance algorithm that compute a Stable* matching: one in which firm-copies propose and the other in which workers propose. These adaptations take into account the specific characteristics of the associated one-to-one market.

The algorithm presented in Table 1 is the adapted firm-copies-proposing deferred acceptance algorithm for the market (P,P¯𝒲)subscript𝑃subscript¯𝑃𝒲(P_{\mathcal{F}},\overline{P}_{\mathcal{W}})( italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT , over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ). While this algorithm follows the same principle as the one introduced by Gale and Shapley (1962), our adaptation includes an important feature: a firm-copy can only make an offer to a worker if no other copy of the same firm is already matched with a more preferred worker. This restriction is essential to prevent envy between firm-copies. Let Owksubscriptsuperscript𝑂𝑘𝑤O^{k}_{w}italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT be the set of firms making a job offer to worker w𝑤witalic_w at stage k𝑘kitalic_k.

Algorithm:
Input A preference profile (P¯𝒲,P)subscript¯𝑃𝒲subscript𝑃(\overline{P}_{\mathcal{W}},P_{\mathcal{F}})( over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT , italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ).
Output A matching λsubscript𝜆\lambda_{\mathcal{F}}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT.
DEFINE: λ0(w)=superscript𝜆0𝑤\lambda^{0}(w)=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_w ) = ∅ and λ0(f)=superscript𝜆0𝑓\lambda^{0}(f)=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_f ) = ∅ for each w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W and each f𝑓f\in\mathcal{F}italic_f ∈ caligraphic_F.
Stage 𝟏1\boldsymbol{1}bold_1 (𝒂)𝒂\boldsymbol{(a)}bold_( bold_italic_a bold_) Each firm-copy f𝑓fitalic_f makes a job offer to the best acceptable worker at Pfsubscript𝑃𝑓P_{f}italic_P start_POSTSUBSCRIPT italic_f end_POSTSUBSCRIPT.
(𝒃)𝒃\boldsymbol{(b)}bold_( bold_italic_b bold_) Each worker w𝑤witalic_w selects, with respect to P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT, the best firm-copy
         among those in Ow1λ0(w)subscriptsuperscript𝑂1𝑤superscript𝜆0𝑤O^{1}_{w}\cup\lambda^{0}(w)italic_O start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_w ), say f~~𝑓\widetilde{f}over~ start_ARG italic_f end_ARG.
Set Rw1=Ow1{f~}subscriptsuperscript𝑅1𝑤subscriptsuperscript𝑂1𝑤~𝑓R^{1}_{w}=O^{1}_{w}\setminus\{\widetilde{f}\}italic_R start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT = italic_O start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∖ { over~ start_ARG italic_f end_ARG } for each w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W. Define
              λ1(fij)={if fijw𝒲Rw1,wiffij=f~.superscript𝜆1subscript𝑓𝑖𝑗casesif subscript𝑓𝑖𝑗subscript𝑤𝒲superscriptsubscript𝑅𝑤1𝑤ifsubscript𝑓𝑖𝑗~𝑓\lambda^{1}(f_{ij})=\begin{cases}\emptyset&\text{if }f_{ij}\in\cup_{w\in% \mathcal{W}}R_{w}^{1},\\ w&\text{if}f_{ij}=\widetilde{f}.\\ \end{cases}italic_λ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = { start_ROW start_CELL ∅ end_CELL start_CELL if italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ ∪ start_POSTSUBSCRIPT italic_w ∈ caligraphic_W end_POSTSUBSCRIPT italic_R start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT , end_CELL end_ROW start_ROW start_CELL italic_w end_CELL start_CELL if italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT = over~ start_ARG italic_f end_ARG . end_CELL end_ROW
Stage k𝑘\boldsymbol{k}bold_italic_k (𝒂)𝒂\boldsymbol{(a)}bold_( bold_italic_a bold_) Each firm-copy fijw~𝒲Rw~k1subscript𝑓𝑖𝑗subscript~𝑤𝒲subscriptsuperscript𝑅𝑘1~𝑤f_{ij}\in\bigcup_{\widetilde{w}\in\mathcal{W}}R^{k-1}_{\widetilde{w}}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ ⋃ start_POSTSUBSCRIPT over~ start_ARG italic_w end_ARG ∈ caligraphic_W end_POSTSUBSCRIPT italic_R start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT over~ start_ARG italic_w end_ARG end_POSTSUBSCRIPT is authorized to make an offer to the best
         acceptable worker w𝑤witalic_w with respect to Pfijsubscript𝑃subscript𝑓𝑖𝑗P_{f_{ij}}italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT who has not previously rejected it,
         if there are no fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT and wwsuperscript𝑤𝑤w^{\prime}\neq witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ≠ italic_w such that λk1(w)=fijsuperscript𝜆𝑘1superscript𝑤subscript𝑓𝑖superscript𝑗\lambda^{k-1}(w^{\prime})=f_{ij^{\prime}}italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ) = italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT and wPfijwsuperscript𝑤subscript𝑃subscript𝑓𝑖𝑗𝑤w^{\prime}P_{f_{ij}}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT italic_w.
         Otherwise, fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT is unauthorized to make an offer at this stage.
(𝒃)𝒃\boldsymbol{(b)}bold_( bold_italic_b bold_) Each worker w𝑤witalic_w selects, with respect to P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT, the best firm-copy
         among those in Owkλk1(w)subscriptsuperscript𝑂𝑘𝑤superscript𝜆𝑘1𝑤O^{k}_{w}\cup\lambda^{k-1}(w)italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w ), say f~~𝑓\widetilde{f}over~ start_ARG italic_f end_ARG.
Set Rwk=Owkλk1(w){f~}subscriptsuperscript𝑅𝑘𝑤subscriptsuperscript𝑂𝑘𝑤superscript𝜆𝑘1𝑤~𝑓R^{k}_{w}=O^{k}_{w}\cup\lambda^{k-1}(w)\setminus\{\widetilde{f}\}italic_R start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT = italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w ) ∖ { over~ start_ARG italic_f end_ARG } for each w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W. Define
              λk(fij)={if fijw𝒲Rwk,if fij is unauthorized to make an offer,wiffij=f~, andλk1(fij)otherwise.superscript𝜆𝑘subscript𝑓𝑖𝑗casesif subscript𝑓𝑖𝑗subscript𝑤𝒲superscriptsubscript𝑅𝑤𝑘if fij is unauthorized to make an offer,𝑤ifsubscript𝑓𝑖𝑗~𝑓, andsuperscript𝜆𝑘1subscript𝑓𝑖𝑗otherwise.\lambda^{k}(f_{ij})=\begin{cases}\emptyset&\text{if }f_{ij}\in\cup_{w\in% \mathcal{W}}R_{w}^{k},\\ \emptyset&\text{if $f_{ij}$ is unauthorized to make an offer,}\\ w&\text{if}f_{ij}=\widetilde{f}\text{, and}\\ \lambda^{k-1}(f_{ij})&\text{otherwise.}\end{cases}italic_λ start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = { start_ROW start_CELL ∅ end_CELL start_CELL if italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ ∪ start_POSTSUBSCRIPT italic_w ∈ caligraphic_W end_POSTSUBSCRIPT italic_R start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT , end_CELL end_ROW start_ROW start_CELL ∅ end_CELL start_CELL if italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT is unauthorized to make an offer, end_CELL end_ROW start_ROW start_CELL italic_w end_CELL start_CELL if italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT = over~ start_ARG italic_f end_ARG , and end_CELL end_ROW start_ROW start_CELL italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) end_CELL start_CELL otherwise. end_CELL end_ROW
         If w𝒲Rwk=subscript𝑤𝒲subscriptsuperscript𝑅𝑘𝑤\bigcup_{w\in\mathcal{W}}R^{k}_{w}=\emptyset⋃ start_POSTSUBSCRIPT italic_w ∈ caligraphic_W end_POSTSUBSCRIPT italic_R start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT = ∅:
                     STOP and let λ=λksubscript𝜆superscript𝜆𝑘\lambda_{\mathcal{F}}=\lambda^{k}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT = italic_λ start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT.
         ELSE:
                     CONTINUE TO STAGE k+1𝑘1k+1italic_k + 1
Table 1: Adapted Firm-copies-proposal Deferred Acceptance algorithm

The algorithm will eventually stop in a finite number of stages, as firms begin by making offers to the top of their preference lists and continue to make offers (if authorized) to less preferred workers as their previous offers are rejected. Given that the set of firms and workers is finite, there will eventually be no more rejections, leading to the cessation of the algorithm. Before presenting that the output of the algorithm is a stable* matching, we illustrate the procedure for Example 1.

Example 1 (continued) Considering the associated one-to-one market previously presented, we applied the adapted firm-copies-proposing deferred acceptance algorithm as follows:

Stage 1: Firm-copies offers are presented in the following table:

f11subscript𝑓11f_{11}italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT f12subscript𝑓12f_{12}italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT f13subscript𝑓13f_{13}italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT f14subscript𝑓14f_{14}italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT f15subscript𝑓15f_{15}italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT f16subscript𝑓16f_{16}italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT f21subscript𝑓21f_{21}italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT f22subscript𝑓22f_{22}italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT f23subscript𝑓23f_{23}italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT f24subscript𝑓24f_{24}italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT f25subscript𝑓25f_{25}italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT f26subscript𝑓26f_{26}italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT
w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT

So, the sets of offers that workers receive in Stage 1 are:

Ow11={f11,f12,f13}superscriptsubscript𝑂subscript𝑤11subscript𝑓11subscript𝑓12subscript𝑓13\displaystyle O_{w_{1}}^{1}=\{f_{11},f_{12},f_{13}\}italic_O start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT }
Ow21={f13,f14,f15}superscriptsubscript𝑂subscript𝑤21subscript𝑓13subscript𝑓14subscript𝑓15\displaystyle O_{w_{2}}^{1}=\{f_{13},f_{14},f_{15}\}italic_O start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT }
Ow31={f21,f22,f23}superscriptsubscript𝑂subscript𝑤31subscript𝑓21subscript𝑓22subscript𝑓23\displaystyle O_{w_{3}}^{1}=\{f_{21},f_{22},f_{23}\}italic_O start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT }
Ow41={f24,f25,f26}superscriptsubscript𝑂subscript𝑤41subscript𝑓24subscript𝑓25subscript𝑓26\displaystyle O_{w_{4}}^{1}=\{f_{24},f_{25},f_{26}\}italic_O start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT }

Workers, according to their preferences P¯¯𝑃\overline{P}over¯ start_ARG italic_P end_ARG, accept an offer and the resulting rejection sets are:

Rw11={f12,f13}superscriptsubscript𝑅subscript𝑤11subscript𝑓12subscript𝑓13\displaystyle R_{w_{1}}^{1}=\{f_{12},f_{13}\}italic_R start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT }
Rw21={f15,f16}superscriptsubscript𝑅subscript𝑤21subscript𝑓15subscript𝑓16\displaystyle R_{w_{2}}^{1}=\{f_{15},f_{16}\}italic_R start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT }
Rw31={f22,f23}superscriptsubscript𝑅subscript𝑤31subscript𝑓22subscript𝑓23\displaystyle R_{w_{3}}^{1}=\{f_{22},f_{23}\}italic_R start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT }
Rw41={f25,f26}superscriptsubscript𝑅subscript𝑤41subscript𝑓25subscript𝑓26\displaystyle R_{w_{4}}^{1}=\{f_{25},f_{26}\}italic_R start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT , italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT }

Thus,

λ1=(f11f12f13f14f15f16f21f22f23f24f25f26w1w2w3w4).superscript𝜆1matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\lambda^{1}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{12}&f_{13}&% f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{1}&\emptyset&\emptyset&w_{2}&\emptyset&\emptyset&w_{3}&\emptyset&\emptyset&% w_{4}&\emptyset&\emptyset\end{pmatrix}.italic_λ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) .

Stage 2 (and henceforth): Each firm-copy fijw𝒲Rw1subscriptfijsubscriptw𝒲subscriptsuperscriptR1wf_{ij}\in\bigcup_{w\in\mathcal{W}}R^{1}_{w}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ ⋃ start_POSTSUBSCRIPT italic_w ∈ caligraphic_W end_POSTSUBSCRIPT italic_R start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT is no longer authorized to make an offer to any w𝒲w𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W. To see this consider, for instance, f12Rw11subscriptf12subscriptsuperscriptR1subscriptw1f_{12}\in R^{1}_{w_{1}}italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT ∈ italic_R start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT. We have that w1=λ(f11)P12wksubscriptw1λsubscriptf11subscriptP12subscriptwkw_{1}=\lambda(f_{11})P_{12}w_{k}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = italic_λ ( italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT italic_k end_POSTSUBSCRIPT, where k{2,4}k24k\in\{2,4\}italic_k ∈ { 2 , 4 }. Therefore, the output of the adapted firm-copies-proposing deferred acceptance algorithm is:

λ=(f11f12f13f14f15f16f21f22f23f24f25f26w1w2w3w4).subscript𝜆matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\lambda_{\mathcal{F}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{1}&\emptyset&\emptyset&w_{2}&\emptyset&\emptyset&w_{3}&\emptyset&\emptyset&% w_{4}&\emptyset&\emptyset\end{pmatrix}.italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) .

\Diamond

The following theorem proves the output of the adapted firm-copies-proposing deferred acceptance algorithm is a stable* matching.

Theorem 1

Let λsubscript𝜆\lambda_{\mathcal{F}}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT be the output of the firm-copies-proposing deferred acceptance algorithm. Then, λsubscript𝜆\lambda_{\mathcal{F}}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT is a stable* matching.

Proof. Assume that λS(M).subscript𝜆superscript𝑆𝑀\lambda_{\mathcal{F}}\notin S^{*}(M).italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ∉ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) . Then, by Remark 1 there are four possible situations:

  1. (i)

    There is w𝒲suchthatP¯wλ(w)𝑤𝒲𝑠𝑢𝑐𝑡𝑎𝑡subscriptbold-¯𝑃𝑤subscript𝜆𝑤\boldsymbol{w\in\mathcal{W}$suchthat$\emptyset\overline{P}_{w}\lambda_{% \mathcal{F}}(w)}bold_italic_w bold_∈ bold_caligraphic_W bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_∅ overbold_¯ start_ARG bold_italic_P end_ARG start_POSTSUBSCRIPT bold_italic_w end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_w bold_). Given that at each stage k𝑘kitalic_k, w𝑤witalic_w selects f~~𝑓\widetilde{f}over~ start_ARG italic_f end_ARG as the best firm-copy among those in Owkλk1(w)subscriptsuperscript𝑂𝑘𝑤superscript𝜆𝑘1𝑤O^{k}_{w}\cup\lambda^{k-1}(w)italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w ) with respect to P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT, and in stage 1 λ0(w)=superscript𝜆0𝑤\lambda^{0}(w)=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_w ) = ∅, we have that λ(w)R¯wsubscript𝜆𝑤subscript¯𝑅𝑤\lambda_{\mathcal{F}}(w)\overline{R}_{w}\emptysetitalic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_w ) over¯ start_ARG italic_R end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∅. Then, this situation does not occur.

  2. (ii)

    There is fijsuchthatPijλ(fij)subscript𝑓𝑖𝑗𝑠𝑢𝑐𝑡𝑎𝑡subscript𝑃𝑖𝑗subscript𝜆subscript𝑓𝑖𝑗\boldsymbol{f_{ij}\in\mathcal{F}$suchthat$\emptyset P_{ij}\lambda_{\mathcal{F}% }(f_{ij})}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_∈ bold_caligraphic_F bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_∅ bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_). Given that at each stage fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT makes an offer, if it is authorized, to an acceptable worker with respect to Pijsubscript𝑃𝑖𝑗P_{ij}italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, we have that λ(fij)Rij.subscript𝜆subscript𝑓𝑖𝑗subscript𝑅𝑖𝑗\lambda_{\mathcal{F}}(f_{ij})R_{ij}\emptyset.italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_R start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅ . Then, this situation does not occur.

  3. (iii)

    There are fij,fijsuchthatλ(fij)Pijλ(fij)Pijsubscript𝑓𝑖𝑗subscript𝑓𝑖superscript𝑗bold-′𝑠𝑢𝑐𝑡𝑎𝑡subscript𝜆subscript𝑓𝑖superscript𝑗bold-′subscript𝑃𝑖𝑗subscript𝜆subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗\boldsymbol{f_{ij},f_{ij^{\prime}}\in\mathcal{F}$suchthat$\lambda_{\mathcal{F}% }(f_{ij^{\prime}})P_{ij}\lambda_{\mathcal{F}}(f_{ij})P_{ij}\emptyset}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_, bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_∈ bold_caligraphic_F bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_) bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_) bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_∅. Assume that there are w,w𝒲𝑤superscript𝑤𝒲w,w^{\prime}\in\mathcal{W}italic_w , italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ caligraphic_W such that w=λ(fij)𝑤subscript𝜆subscript𝑓𝑖𝑗w=\lambda_{\mathcal{F}}(f_{ij})italic_w = italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) and w=λ(fij).superscript𝑤subscript𝜆subscript𝑓𝑖superscript𝑗w^{\prime}=\lambda_{\mathcal{F}}(f_{ij^{\prime}}).italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) . Assume also that there are two stages t,t𝑡superscript𝑡t,t^{\prime}italic_t , italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT such that tsuperscript𝑡t^{\prime}italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT is the stage where fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT is assigned to wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT (λt(fij)=wsuperscript𝜆superscript𝑡subscript𝑓𝑖superscript𝑗superscript𝑤\lambda^{t^{\prime}}(f_{ij^{\prime}})=w^{\prime}italic_λ start_POSTSUPERSCRIPT italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) = italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT and λt1(fij)wsuperscript𝜆superscript𝑡1subscript𝑓𝑖superscript𝑗superscript𝑤\lambda^{t^{\prime}-1}(f_{ij^{\prime}})\neq w^{\prime}italic_λ start_POSTSUPERSCRIPT italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) ≠ italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT) and such that t𝑡titalic_t is the stage where fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT is assigned to w𝑤witalic_w (λt(fij)=wsuperscript𝜆𝑡subscript𝑓𝑖𝑗𝑤\lambda^{t}(f_{ij})=witalic_λ start_POSTSUPERSCRIPT italic_t end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_w and λt1(fij)wsuperscript𝜆𝑡1subscript𝑓𝑖𝑗𝑤\lambda^{t-1}(f_{ij})\neq witalic_λ start_POSTSUPERSCRIPT italic_t - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) ≠ italic_w). We have to consider two subcases.

    𝒕<𝒕::superscript𝒕bold-′𝒕absent\boldsymbol{t^{\prime}<t}:bold_italic_t start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_< bold_italic_t : Since wPijwsuperscript𝑤subscript𝑃𝑖𝑗𝑤w^{\prime}P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w, we have that fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT is unauthorized to make an offer to w𝑤witalic_w, so it is not possible that λ(fij)=w.subscript𝜆subscript𝑓𝑖𝑗𝑤\lambda_{\mathcal{F}}(f_{ij})=w.italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_w .

    𝒕𝒕superscript𝒕bold-′𝒕\boldsymbol{t^{\prime}\geq t}bold_italic_t start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_≥ bold_italic_t: Note that since w𝑤witalic_w is matched with fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT before than wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT and the fact that wPijwsuperscript𝑤subscript𝑃𝑖𝑗𝑤w^{\prime}P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w implies that wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT reject fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT in an earlier step that t𝑡titalic_t, say t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG. In this case, we claim that there is a firm-copy fi~ksubscript𝑓~𝑖𝑘f_{\widetilde{i}k}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT that is not a copy of fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT such that w=λt~(fi~k)superscript𝑤superscript𝜆~𝑡subscript𝑓~𝑖𝑘w^{\prime}=\lambda^{\widetilde{t}}(f_{\widetilde{i}k})italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUPERSCRIPT over~ start_ARG italic_t end_ARG end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT ) and fi~kP¯wfi,jsubscript𝑓~𝑖𝑘subscript¯𝑃superscript𝑤subscript𝑓𝑖𝑗f_{\widetilde{i}k}\overline{P}_{w^{\prime}}f_{i,j}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i , italic_j end_POSTSUBSCRIPT. Since wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT reject fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT at stage t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG, such a firm-copy exists that is temporarily matched to wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT at stage t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG, and wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT prefers this firm-copy over fij.subscript𝑓𝑖𝑗f_{ij}.italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT . Moreover, since t~t~𝑡𝑡\widetilde{t}\leq tover~ start_ARG italic_t end_ARG ≤ italic_t, we have that fi~ksubscript𝑓~𝑖𝑘f_{\widetilde{i}k}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT is not a copy of fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, otherwise fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT will not be authorized to make an offer to w𝑤witalic_w. Thus, in this case, the claim holds. Note that this contradicts the definition of workers’ preferences P¯¯𝑃\overline{P}over¯ start_ARG italic_P end_ARG item (1), since fijP¯wfi~kP¯wfi,j.subscript𝑓𝑖superscript𝑗subscript¯𝑃superscript𝑤subscript𝑓~𝑖𝑘subscript¯𝑃superscript𝑤subscript𝑓𝑖𝑗f_{ij^{\prime}}\overline{P}_{w^{\prime}}f_{\widetilde{i}k}\overline{P}_{w^{% \prime}}f_{i,j}.italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i , italic_j end_POSTSUBSCRIPT .

    Then, by the two cases analyzed, this situation does not occur.

  4. (iv)

    There is a pair (fij,w)×𝒲subscript𝑓𝑖𝑗𝑤𝒲\boldsymbol{(f_{ij},w)\in\mathcal{F}\times\mathcal{W}}bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_, bold_italic_w bold_) bold_∈ bold_caligraphic_F bold_× bold_caligraphic_W such that fijP¯wλ(w)subscript𝑓𝑖𝑗subscriptbold-¯𝑃𝑤subscript𝜆𝑤\boldsymbol{f_{ij}\overline{P}_{w}\lambda_{\mathcal{F}}(w)}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT overbold_¯ start_ARG bold_italic_P end_ARG start_POSTSUBSCRIPT bold_italic_w end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_w bold_) and wPijλ(fij)𝑤subscript𝑃𝑖𝑗subscript𝜆subscript𝑓𝑖superscript𝑗bold-′\boldsymbol{wP_{ij}\lambda_{\mathcal{F}}(f_{ij^{\prime}})}bold_italic_w bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_F end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_) for each jJisuperscript𝑗bold-′subscript𝐽𝑖\boldsymbol{j^{\prime}\in J_{i}}bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_∈ bold_italic_J start_POSTSUBSCRIPT bold_italic_i end_POSTSUBSCRIPT. If wPijλ(fij)𝑤subscript𝑃𝑖𝑗subscript𝜆subscript𝑓𝑖superscript𝑗wP_{ij}\lambda_{\mathcal{F}}(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJisuperscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT, then each fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT has made an offer to w𝑤witalic_w in a previous stage and it was rejected because w𝑤witalic_w was temporarily matched with a better firm-copy that is not a copy of fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT, otherwise fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT is unauthorized to make an offer. Then, λ(w)P¯wfijsubscript𝜆𝑤subscript¯𝑃𝑤subscript𝑓𝑖superscript𝑗\lambda_{\mathcal{F}}(w)\overline{P}_{w}f_{ij^{\prime}}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_w ) over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT for each jJi.superscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}.italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT . Then, this situation does not occur.

    Therefore, since none of the four cases hold, we have that λS(M).subscript𝜆superscript𝑆𝑀\lambda_{\mathcal{F}}\in S^{*}(M).italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) .

\square

The algorithm presented in Table 2 is the adapted worker-proposing deferred acceptance algorithm for the market (P¯𝒲,P)subscript¯𝑃𝒲subscript𝑃(\overline{P}_{\mathcal{W}},P_{\mathcal{F}})( over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT , italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ). This adaptation also ensures that the output is envy-free among firm-copies. We say that a worker makes a valid offer to a firm-copy if the acceptance of the worker’s proposal by that firm-copy does not generate envy among the other copies.

Let Ofksubscriptsuperscript𝑂𝑘𝑓O^{k}_{f}italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f end_POSTSUBSCRIPT be the set of workers making a job offer to firm f𝑓fitalic_f at stage k𝑘kitalic_k.

Algorithm:
Input A preference profile (P¯𝒲,P)subscript¯𝑃𝒲subscript𝑃(\overline{P}_{\mathcal{W}},P_{\mathcal{F}})( over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT , italic_P start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ).
Output A matching λ𝒲subscript𝜆𝒲\lambda_{\mathcal{W}}italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT.
DEFINE: λ0(w)=superscript𝜆0𝑤\lambda^{0}(w)=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_w ) = ∅ and λ0(fij)=superscript𝜆0subscript𝑓𝑖𝑗\lambda^{0}(f_{ij})=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = ∅ for each w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W and each fijsubscript𝑓𝑖𝑗f_{ij}\in\mathcal{F}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F.
Stage 𝟏1\boldsymbol{1}bold_1 (𝒂)𝒂\boldsymbol{(a)}bold_( bold_italic_a bold_) Each worker w𝑤witalic_w makes an offer to the best acceptable firm-copy at P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT.
(𝒃)𝒃\boldsymbol{(b)}bold_( bold_italic_b bold_) Each firm-copy fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT selects, with respect to Pfijsubscript𝑃subscript𝑓𝑖𝑗P_{f_{ij}}italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT, the best worker
         among those in Ofij1λ0(fij)subscriptsuperscript𝑂1subscript𝑓𝑖𝑗superscript𝜆0subscript𝑓𝑖𝑗O^{1}_{f_{ij}}\cup\lambda^{0}(f_{ij})italic_O start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), say w~fijsubscript~𝑤subscript𝑓𝑖𝑗\widetilde{w}_{f_{ij}}over~ start_ARG italic_w end_ARG start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT.
Set Rfij1=Ofij1{w~fij}subscriptsuperscript𝑅1subscript𝑓𝑖𝑗subscriptsuperscript𝑂1subscript𝑓𝑖𝑗subscript~𝑤subscript𝑓𝑖𝑗R^{1}_{f_{ij}}=O^{1}_{f_{ij}}\setminus\{\widetilde{w}_{f_{ij}}\}italic_R start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT = italic_O start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∖ { over~ start_ARG italic_w end_ARG start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT } for each fijsubscript𝑓𝑖𝑗f_{ij}\in\mathcal{F}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F. Define
              λ1(w)={if wfijRfij1,fijif w=w~fij.superscript𝜆1𝑤casesif 𝑤subscriptsubscript𝑓𝑖𝑗superscriptsubscript𝑅subscript𝑓𝑖𝑗1subscript𝑓𝑖𝑗if 𝑤subscript~𝑤subscript𝑓𝑖𝑗\lambda^{1}(w)=\begin{cases}\emptyset&\text{if }w\in\cup_{f_{ij}\in\mathcal{F}% }R_{f_{ij}}^{1},\\ f_{ij}&\text{if }w=\widetilde{w}_{f_{ij}}.\\ \end{cases}italic_λ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT ( italic_w ) = { start_ROW start_CELL ∅ end_CELL start_CELL if italic_w ∈ ∪ start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F end_POSTSUBSCRIPT italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT , end_CELL end_ROW start_ROW start_CELL italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_CELL start_CELL if italic_w = over~ start_ARG italic_w end_ARG start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT . end_CELL end_ROW
Stage k𝑘\boldsymbol{k}bold_italic_k (𝒂)𝒂\boldsymbol{(a)}bold_( bold_italic_a bold_) Each worker wfijRfijk1𝑤subscriptsubscript𝑓𝑖𝑗subscriptsuperscript𝑅𝑘1subscript𝑓𝑖𝑗w\in\bigcup_{f_{ij}\in\mathcal{F}}R^{k-1}_{f_{ij}}italic_w ∈ ⋃ start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F end_POSTSUBSCRIPT italic_R start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT make an new offer to the best
         acceptable firm-copy fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT with respect to P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT who has not previously rejected it.
(𝒃)𝒃\boldsymbol{(b)}bold_( bold_italic_b bold_) For each firm-copy fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, define the set of “Valid Offers”
O~ijk=Oijk{wOijk:w𝒲 with w=λk1(fij)Pfijw for some jJi}superscriptsubscript~𝑂𝑖𝑗𝑘superscriptsubscript𝑂𝑖𝑗𝑘conditional-set𝑤superscriptsubscript𝑂𝑖𝑗𝑘superscript𝑤𝒲 with superscript𝑤superscript𝜆𝑘1subscript𝑓𝑖superscript𝑗subscript𝑃subscript𝑓𝑖𝑗𝑤 for some superscript𝑗subscript𝐽𝑖\widetilde{O}_{ij}^{k}=O_{ij}^{k}\setminus\{w\in O_{ij}^{k}:\exists w^{\prime}% \in\mathcal{W}\text{ with }w^{\prime}=\lambda^{k-1}(f_{ij^{\prime}})P_{f_{ij}}% w\text{ for some }j^{\prime}\in J_{i}\}over~ start_ARG italic_O end_ARG start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT = italic_O start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT ∖ { italic_w ∈ italic_O start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT : ∃ italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ caligraphic_W with italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT italic_w for some italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT }.
Each firm fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT selects the best worker, with respect to Pfijsubscript𝑃subscript𝑓𝑖𝑗P_{f_{ij}}italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT
         among those in O~fijkλk1(fij)subscriptsuperscript~𝑂𝑘subscript𝑓𝑖𝑗superscript𝜆𝑘1subscript𝑓𝑖𝑗\widetilde{O}^{k}_{f_{ij}}\cup\lambda^{k-1}(f_{ij})over~ start_ARG italic_O end_ARG start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), say w~ijsuperscript~𝑤𝑖𝑗\widetilde{w}^{ij}over~ start_ARG italic_w end_ARG start_POSTSUPERSCRIPT italic_i italic_j end_POSTSUPERSCRIPT.
Set Rfijk=Ofijkλk1(fij){w~ij}subscriptsuperscript𝑅𝑘subscript𝑓𝑖𝑗subscriptsuperscript𝑂𝑘subscript𝑓𝑖𝑗superscript𝜆𝑘1subscript𝑓𝑖𝑗superscript~𝑤𝑖𝑗R^{k}_{f_{ij}}=O^{k}_{f_{ij}}\cup\lambda^{k-1}(f_{ij})\setminus\{\widetilde{w}% ^{ij}\}italic_R start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT = italic_O start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) ∖ { over~ start_ARG italic_w end_ARG start_POSTSUPERSCRIPT italic_i italic_j end_POSTSUPERSCRIPT } for each fijsubscript𝑓𝑖𝑗f_{ij}\in\mathcal{F}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F. Define
              λk(w)={if wfijRfijk,fijif w=w~ij, andλk1(w)otherwise.superscript𝜆𝑘𝑤casesif 𝑤subscriptsubscript𝑓𝑖𝑗superscriptsubscript𝑅subscript𝑓𝑖𝑗𝑘subscript𝑓𝑖𝑗if 𝑤superscript~𝑤𝑖𝑗, andsuperscript𝜆𝑘1𝑤otherwise.\lambda^{k}(w)=\begin{cases}\emptyset&\text{if }w\in\cup_{f_{ij}\in\mathcal{F}% }R_{f_{ij}}^{k},\\ f_{ij}&\text{if }w=\widetilde{w}^{ij}\text{, and}\\ \lambda^{k-1}(w)&\text{otherwise.}\end{cases}italic_λ start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT ( italic_w ) = { start_ROW start_CELL ∅ end_CELL start_CELL if italic_w ∈ ∪ start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F end_POSTSUBSCRIPT italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT , end_CELL end_ROW start_ROW start_CELL italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_CELL start_CELL if italic_w = over~ start_ARG italic_w end_ARG start_POSTSUPERSCRIPT italic_i italic_j end_POSTSUPERSCRIPT , and end_CELL end_ROW start_ROW start_CELL italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w ) end_CELL start_CELL otherwise. end_CELL end_ROW
         If fijRfijk=subscriptsubscript𝑓𝑖𝑗subscriptsuperscript𝑅𝑘subscript𝑓𝑖𝑗\bigcup_{f_{ij}\in\mathcal{F}}R^{k}_{f_{ij}}=\emptyset⋃ start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∈ caligraphic_F end_POSTSUBSCRIPT italic_R start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT = ∅:
                     STOP and let λ𝒲=λksubscript𝜆𝒲superscript𝜆𝑘\lambda_{\mathcal{W}}=\lambda^{k}italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = italic_λ start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT.
         ELSE:
                     CONTINUE TO STAGE k+1𝑘1k+1italic_k + 1
Table 2: Adapted Worker-proposal Deferred Acceptance algorithm

Before proving that the output of the algorithm is a stable* matching, we first illustrate the procedure with Example 1.

Example 1 (continued) Considering the associated one-to-one market previously presented, we applied the adapted worker-proposing deferred acceptance algorithm as follows:

Stage 1: The workers’ offers are presented in the following table:

w1subscript𝑤1w_{1}italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT w2subscript𝑤2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT w3subscript𝑤3w_{3}italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT w4subscript𝑤4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT
f21subscript𝑓21f_{21}italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT f21subscript𝑓21f_{21}italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT f11subscript𝑓11f_{11}italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT f11subscript𝑓11f_{11}italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT

So, the sets of offers that firm-copies receive in stage 1 are: Of1,11={w3,w4}superscriptsubscript𝑂subscript𝑓111subscript𝑤3subscript𝑤4O_{f_{1,1}}^{1}=\{w_{3},w_{4}\}italic_O start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 1 , 1 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT } and Of211={w1,w2}.superscriptsubscript𝑂subscript𝑓211subscript𝑤1subscript𝑤2O_{f_{21}}^{1}=\{w_{1},w_{2}\}.italic_O start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = { italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT } . Firm-copies, according to their preferences P𝑃Pitalic_P, accept the offers, and the resulting rejection sets are Rf111=w4superscriptsubscript𝑅subscript𝑓111subscript𝑤4R_{f_{11}}^{1}={w_{4}}italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT and Rf211=w2superscriptsubscript𝑅subscript𝑓211subscript𝑤2R_{f_{21}}^{1}={w_{2}}italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT. Thus,

λ1=(f11f12f13f14f15f16f21f22f23f24f25f26w3w2).superscript𝜆1matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤3subscript𝑤2\lambda^{1}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{12}&f_{13}&% f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{3}&\emptyset&\emptyset&\emptyset&\emptyset&\emptyset&w_{2}&\emptyset&% \emptyset&\emptyset&\emptyset&\emptyset\end{pmatrix}.italic_λ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) .

Stage 2: Each worker w2,w4Rf111Rf211subscriptw2subscriptw4superscriptsubscriptRsubscriptf111superscriptsubscriptRsubscriptf211{w_{2},w_{4}}\in R_{f_{11}}^{1}\cup R_{f_{21}}^{1}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT ∈ italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT ∪ italic_R start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT submits a new offer. Worker w2subscriptw2w_{2}italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT offers to firm-copy f22subscriptf22f_{22}italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT, while w4subscriptw4w_{4}italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT offers to firm-copy f1,2subscriptf12f_{1,2}italic_f start_POSTSUBSCRIPT 1 , 2 end_POSTSUBSCRIPT. Since both offers are valid, and firm-copies f12subscriptf12f_{12}italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT and f22subscriptf22f_{22}italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT accept them without generating any further rejections, the algorithm stops. Thus,

λ𝒲=(f11f12f13f14f15f16f21f22f23f24f25f26w3w4w2w1).subscript𝜆𝒲matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤3subscript𝑤4subscript𝑤2subscript𝑤1\lambda_{\mathcal{W}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{3}&w_{4}&\emptyset&\emptyset&\emptyset&\emptyset&w_{2}&w_{1}&\emptyset&% \emptyset&\emptyset&\emptyset\end{pmatrix}.italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) .

\Diamond

The following theorem shows that the output of the adapted worker-proposing deferred acceptance algorithm is a stable* matching. The proof follows the same spirit as the proof of Theorem 1, adapted for the case where workers are the ones making the offers.

Theorem 2

Let λ𝒲subscript𝜆𝒲\lambda_{\mathcal{W}}italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT be the output of the worker-proposing deferred acceptance algorithm. Then, λ𝒲subscript𝜆𝒲\lambda_{\mathcal{W}}italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT is a stable* matching.

Proof. Assume that λ𝒲S(M).subscript𝜆𝒲superscript𝑆𝑀\lambda_{\mathcal{W}}\notin S^{*}(M).italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ∉ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) . Then, by Remark 1 there are four possible situations:

  1. (i)

    There is fijsuchthatPijλ𝒲(fij)subscript𝑓𝑖𝑗𝑠𝑢𝑐𝑡𝑎𝑡subscript𝑃𝑖𝑗subscript𝜆𝒲subscript𝑓𝑖𝑗\boldsymbol{f_{ij}\in\mathcal{F}$suchthat$\emptyset P_{ij}\lambda_{\mathcal{W}% }(f_{ij})}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_∈ bold_caligraphic_F bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_∅ bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_W end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_). Given that at each stage k𝑘kitalic_k, fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT selects w~ijsuperscript~𝑤𝑖𝑗\widetilde{w}^{ij}over~ start_ARG italic_w end_ARG start_POSTSUPERSCRIPT italic_i italic_j end_POSTSUPERSCRIPT as the best worker among those in O~fijkλk1(w)subscriptsuperscript~𝑂𝑘subscript𝑓𝑖𝑗superscript𝜆𝑘1𝑤\widetilde{O}^{k}_{f_{ij}}\cup\lambda^{k-1}(w)over~ start_ARG italic_O end_ARG start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∪ italic_λ start_POSTSUPERSCRIPT italic_k - 1 end_POSTSUPERSCRIPT ( italic_w ) with respect to Pijsubscript𝑃𝑖𝑗P_{ij}italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, and in stage 1 λ0(fij)=superscript𝜆0subscript𝑓𝑖𝑗\lambda^{0}(f_{ij})=\emptysetitalic_λ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = ∅, we have that λ𝒲(fij)Rijsubscript𝜆𝒲subscript𝑓𝑖𝑗subscript𝑅𝑖𝑗\lambda_{\mathcal{W}}(f_{ij})R_{ij}\emptysetitalic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_R start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅. Then, this situation does not occur.

  2. (ii)

    There is w𝒲suchthatP¯wλ(w)𝑤𝒲𝑠𝑢𝑐𝑡𝑎𝑡subscriptbold-¯𝑃𝑤𝜆𝑤\boldsymbol{w\in\mathcal{W}$suchthat$\emptyset\overline{P}_{w}\lambda(w)}bold_italic_w bold_∈ bold_caligraphic_W bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_∅ overbold_¯ start_ARG bold_italic_P end_ARG start_POSTSUBSCRIPT bold_italic_w end_POSTSUBSCRIPT bold_italic_λ bold_( bold_italic_w bold_). Given that at each stage w𝑤witalic_w makes an offer, to an acceptable firm with respect to P¯wsubscript¯𝑃𝑤\overline{P}_{w}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT, we have that λ(w)𝒲R¯w.𝜆subscript𝑤𝒲subscript¯𝑅𝑤\lambda(w)_{\mathcal{W}}\overline{R}_{w}\emptyset.italic_λ ( italic_w ) start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT over¯ start_ARG italic_R end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∅ . Then, this situation does not occur.

  3. (iii)

    There are fij,fijsuchthatλ𝒲(fij)Pijλ𝒲(fij)Pijsubscript𝑓𝑖𝑗subscript𝑓𝑖superscript𝑗bold-′𝑠𝑢𝑐𝑡𝑎𝑡subscript𝜆𝒲subscript𝑓𝑖superscript𝑗bold-′subscript𝑃𝑖𝑗subscript𝜆𝒲subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗\boldsymbol{f_{ij},f_{ij^{\prime}}\in\mathcal{F}$suchthat$\lambda_{\mathcal{W}% }(f_{ij^{\prime}})P_{ij}\lambda_{\mathcal{W}}(f_{ij})P_{ij}\emptyset}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_, bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_∈ bold_caligraphic_F bold_italic_s bold_italic_u bold_italic_c bold_italic_h bold_italic_t bold_italic_h bold_italic_a bold_italic_t bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_W end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_) bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_W end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_) bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_∅. Assume that there are w,w𝒲𝑤superscript𝑤𝒲w,w^{\prime}\in\mathcal{W}italic_w , italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ caligraphic_W such that w=λ𝒲(fij)𝑤subscript𝜆𝒲subscript𝑓𝑖𝑗w=\lambda_{\mathcal{W}}(f_{ij})italic_w = italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) and w=λ𝒲(fij).superscript𝑤subscript𝜆𝒲subscript𝑓𝑖superscript𝑗w^{\prime}=\lambda_{\mathcal{W}}(f_{ij^{\prime}}).italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) . Assume also that there are two stages t,t𝑡superscript𝑡t,t^{\prime}italic_t , italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT such that tsuperscript𝑡t^{\prime}italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT is the stage where fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT is assigned to wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT (λt(fij)=wsuperscript𝜆superscript𝑡subscript𝑓𝑖superscript𝑗superscript𝑤\lambda^{t^{\prime}}(f_{ij^{\prime}})=w^{\prime}italic_λ start_POSTSUPERSCRIPT italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) = italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT and λt1(fij)wsuperscript𝜆superscript𝑡1subscript𝑓𝑖superscript𝑗superscript𝑤\lambda^{t^{\prime}-1}(f_{ij^{\prime}})\neq w^{\prime}italic_λ start_POSTSUPERSCRIPT italic_t start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) ≠ italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT) and such that t𝑡titalic_t is the stage where fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT is assigned to w𝑤witalic_w (λt(fij)=wsuperscript𝜆𝑡subscript𝑓𝑖𝑗𝑤\lambda^{t}(f_{ij})=witalic_λ start_POSTSUPERSCRIPT italic_t end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_w and λt1(fij)wsuperscript𝜆𝑡1subscript𝑓𝑖𝑗𝑤\lambda^{t-1}(f_{ij})\neq witalic_λ start_POSTSUPERSCRIPT italic_t - 1 end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) ≠ italic_w). There are two subcases to consider.

    𝒕<𝒕::superscript𝒕bold-′𝒕absent\boldsymbol{t^{\prime}<t}:bold_italic_t start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_< bold_italic_t : Since wPijwsuperscript𝑤subscript𝑃𝑖𝑗𝑤w^{\prime}P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w, we have that w𝑤witalic_w is not a valid offer for fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, so it is not possible that λ𝒲(fij)=w.subscript𝜆𝒲subscript𝑓𝑖𝑗𝑤\lambda_{\mathcal{W}}(f_{ij})=w.italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_w .

    𝒕𝒕superscript𝒕bold-′𝒕\boldsymbol{t^{\prime}\geq t}bold_italic_t start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_≥ bold_italic_t: Note that since w𝑤witalic_w is matched with fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT before that wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT and the fact that wPijwsuperscript𝑤subscript𝑃𝑖𝑗𝑤w^{\prime}P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w implies that wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT reject fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT in an earlier step that t𝑡titalic_t, say t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG. In this case, we claim that there is a firm fi~ksubscript𝑓~𝑖𝑘f_{\widetilde{i}k}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT that is not a copy of fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT such that w=λt~(fi~k)superscript𝑤superscript𝜆~𝑡subscript𝑓~𝑖𝑘w^{\prime}=\lambda^{\widetilde{t}}(f_{\widetilde{i}k})italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUPERSCRIPT over~ start_ARG italic_t end_ARG end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT ) and fi~kP¯wfi,jsubscript𝑓~𝑖𝑘subscript¯𝑃superscript𝑤subscript𝑓𝑖𝑗f_{\widetilde{i}k}\overline{P}_{w^{\prime}}f_{i,j}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i , italic_j end_POSTSUBSCRIPT. Since wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT reject fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT at stage t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG, such a firm-copy exists and is temporarily matched to wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT at stage t~~𝑡\widetilde{t}over~ start_ARG italic_t end_ARG, and wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT prefers this firm-copy over fij.subscript𝑓𝑖𝑗f_{ij}.italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT . Moreover, since t~t~𝑡𝑡\widetilde{t}\leq tover~ start_ARG italic_t end_ARG ≤ italic_t, we have that fi~ksubscript𝑓~𝑖𝑘f_{\widetilde{i}k}italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT is not a copy of fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, otherwise w𝑤witalic_w is not a valid offer for fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT. Thus, in this case, the claim holds. Note that this contradicts the definition of workers’ preferences P¯¯𝑃\overline{P}over¯ start_ARG italic_P end_ARG item (1), since fijP¯wfi~kP¯wfi,j.subscript𝑓𝑖superscript𝑗subscript¯𝑃superscript𝑤subscript𝑓~𝑖𝑘subscript¯𝑃superscript𝑤subscript𝑓𝑖𝑗f_{ij^{\prime}}\overline{P}_{w^{\prime}}f_{\widetilde{i}k}\overline{P}_{w^{% \prime}}f_{i,j}.italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT over~ start_ARG italic_i end_ARG italic_k end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i , italic_j end_POSTSUBSCRIPT .

    Then, by the two cases analyzed, this situation does not occur.

  4. (iv)

    There is a pair (fij,w)×𝒲subscript𝑓𝑖𝑗𝑤𝒲\boldsymbol{(f_{ij},w)\in\mathcal{F}\times\mathcal{W}}bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_, bold_italic_w bold_) bold_∈ bold_caligraphic_F bold_× bold_caligraphic_W such that fijP¯wλ𝒲(w)subscript𝑓𝑖𝑗subscriptbold-¯𝑃𝑤subscript𝜆𝒲𝑤\boldsymbol{f_{ij}\overline{P}_{w}\lambda_{\mathcal{W}}(w)}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT overbold_¯ start_ARG bold_italic_P end_ARG start_POSTSUBSCRIPT bold_italic_w end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_W end_POSTSUBSCRIPT bold_( bold_italic_w bold_) and wPijλ𝒲(fij)𝑤subscript𝑃𝑖𝑗subscript𝜆𝒲subscript𝑓𝑖superscript𝑗bold-′\boldsymbol{wP_{ij}\lambda_{\mathcal{W}}(f_{ij^{\prime}})}bold_italic_w bold_italic_P start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT bold_italic_λ start_POSTSUBSCRIPT bold_caligraphic_W end_POSTSUBSCRIPT bold_( bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT bold_) for each jJisuperscript𝑗bold-′subscript𝐽𝑖\boldsymbol{j^{\prime}\in J_{i}}bold_italic_j start_POSTSUPERSCRIPT bold_′ end_POSTSUPERSCRIPT bold_∈ bold_italic_J start_POSTSUBSCRIPT bold_italic_i end_POSTSUBSCRIPT. There are two subcases to consider.

    If w𝑤\boldsymbol{w}bold_italic_w makes a non-valid offer to fijsubscript𝑓𝑖𝑗\boldsymbol{f_{ij}}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT at some step: then, there is a stage t𝑡titalic_t and a worker wsuperscript𝑤w^{\prime}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT such that w=λt(fij)Pijwsuperscript𝑤superscript𝜆𝑡subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗𝑤w^{\prime}=\lambda^{t}(f_{ij})P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUPERSCRIPT italic_t end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w, contradicting that wPfijλ𝒲(fij)𝑤subscript𝑃subscript𝑓𝑖𝑗subscript𝜆𝒲subscript𝑓𝑖superscript𝑗wP_{f_{ij}}\lambda_{\mathcal{W}}(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJi.superscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}.italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT .

    If w𝑤\boldsymbol{w}bold_italic_w makes a valid offer to fijsubscript𝑓𝑖𝑗\boldsymbol{f_{ij}}bold_italic_f start_POSTSUBSCRIPT bold_italic_i bold_italic_j end_POSTSUBSCRIPT at some step: Then, w𝑤witalic_w was rejected by fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT at some stage k𝑘kitalic_k. Thus, there is w𝒲superscript𝑤𝒲w^{\prime}\in\mathcal{W}italic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ caligraphic_W such that w=λk(fij)Pijwsuperscript𝑤superscript𝜆𝑘subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗𝑤w^{\prime}=\lambda^{k}(f_{ij})P_{ij}witalic_w start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = italic_λ start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_w, contradicting that wPijλ𝒲(fij)𝑤subscript𝑃𝑖𝑗subscript𝜆𝒲subscript𝑓𝑖superscript𝑗wP_{ij}\lambda_{\mathcal{W}}(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJi.superscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}.italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT .

Therefore, since none of the four cases hold, we have that λ𝒲S(M).subscript𝜆𝒲superscript𝑆𝑀\lambda_{\mathcal{W}}\in S^{*}(M).italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) .   \square

3.2 The Isomorphism between markets

In this subsection, we prove that there is an isomorphism between the set of stable matchings in a many-to-one market with substitutable preferences and the set of stable* matchings of the associated one-to-one market, which is derived from the Aizermann-Malishevski decomposition.

Theorem 3

There is an isomorphism between S(M)superscript𝑆𝑀S^{*}(M)italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) and S()𝑆S({{\mathcal{M}}})italic_S ( caligraphic_M ).

Proof. Given λS(M)𝜆superscript𝑆𝑀\lambda\in S^{*}\left(M\right)italic_λ ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ), define T(λ)=μ𝑇𝜆𝜇T(\lambda)=\muitalic_T ( italic_λ ) = italic_μ as

μ(φi)=jJiλ(fij)𝜇subscript𝜑𝑖subscript𝑗subscript𝐽𝑖𝜆subscript𝑓𝑖𝑗\mu(\varphi_{i})=\bigcup\limits_{j\in J_{i}}\lambda(f_{ij})italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) = ⋃ start_POSTSUBSCRIPT italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT )

for each φiΦ.subscript𝜑𝑖Φ\varphi_{i}\in\Phi.italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ∈ roman_Φ . Moreover, if wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ), then μ(w)=φi𝜇𝑤subscript𝜑𝑖\mu(w)=\varphi_{i}italic_μ ( italic_w ) = italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT.

Similarly, if μS(),𝜇𝑆\mu\in S\left({{{\mathcal{M}}}}\right),italic_μ ∈ italic_S ( caligraphic_M ) , define T1(μ)=λsuperscript𝑇1𝜇𝜆T^{-1}(\mu)=\lambdaitalic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_μ ) = italic_λ as λ(w)=fij𝜆𝑤subscript𝑓𝑖𝑗\lambda(w)=f_{ij}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT for w=maxμ(φi)Pij𝑤subscript𝜇subscript𝜑𝑖subscript𝑃𝑖𝑗w=\max_{\mu(\varphi_{i})}P_{ij}italic_w = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT such that there is no j<jsuperscript𝑗𝑗j^{\prime}<jitalic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT < italic_j with w=maxμ(φi)Pij𝑤subscript𝜇subscript𝜑𝑖subscript𝑃𝑖superscript𝑗w=\max_{\mu(\varphi_{i})}P_{ij^{\prime}}italic_w = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. Moreover, if λ(w)=fij𝜆𝑤subscript𝑓𝑖𝑗\lambda(w)=f_{ij}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, then λ(fij)=w.𝜆subscript𝑓𝑖𝑗𝑤\lambda(f_{ij})=w.italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_w .

Formally we need to show that,

  1. (i)

    Given λS(M),thenT(λ)=μS()formulae-sequence𝜆superscript𝑆bold-∗𝑀𝑡𝑒𝑛𝑇𝜆𝜇𝑆\boldsymbol{\lambda\in S^{\ast}\left(M\right)$,then$T(\lambda)=\mu\in S\left(% \mathcal{M}\right)}bold_italic_λ bold_∈ bold_italic_S start_POSTSUPERSCRIPT bold_∗ end_POSTSUPERSCRIPT bold_( bold_italic_M bold_) bold_, bold_italic_t bold_italic_h bold_italic_e bold_italic_n bold_italic_T bold_( bold_italic_λ bold_) bold_= bold_italic_μ bold_∈ bold_italic_S bold_( bold_caligraphic_M bold_). We prove this implication by proving the following two claims.

    Claim 1

    μ=T(λ)𝜇𝑇𝜆\mu=T(\lambda)italic_μ = italic_T ( italic_λ ) is a many-to-one matching.

    By definition of T𝑇Titalic_T, for each w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W, wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) implies that there is jJi𝑗subscript𝐽𝑖j\in J_{i}italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT such that w=λ(fij)𝑤𝜆subscript𝑓𝑖𝑗w=\lambda(f_{ij})italic_w = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ). Assume μ𝜇\muitalic_μ is not a matching, then there is φisubscript𝜑superscript𝑖\varphi_{i^{\prime}}italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT with iisuperscript𝑖𝑖i^{\prime}\neq iitalic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ≠ italic_i such that wμ(φi)𝑤𝜇subscript𝜑superscript𝑖w\in\mu(\varphi_{i^{\prime}})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ). So, there is jJisuperscript𝑗subscript𝐽superscript𝑖j^{\prime}\in J_{i^{\prime}}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT such that w=λ(fij)𝑤𝜆subscript𝑓superscript𝑖superscript𝑗w=\lambda(f_{i^{\prime}j^{\prime}})italic_w = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ). Thus, w=λ(fij)=λ(fij)𝑤𝜆subscript𝑓𝑖𝑗𝜆subscript𝑓superscript𝑖superscript𝑗w=\lambda(f_{ij})=\lambda(f_{i^{\prime}j^{\prime}})italic_w = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) which contradicts λ𝜆\lambdaitalic_λ being a one-to-one matching.

    The construction of T𝑇Titalic_T ensures the bilateral nature of the assignment.

    Claim 2

    μ=T(λ)S(M)𝜇𝑇𝜆superscript𝑆𝑀\mu=T(\lambda)\in S^{\ast}\left(M\right)italic_μ = italic_T ( italic_λ ) ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ).

    By way of contradiction, assume μS()𝜇𝑆\mu\notin S(\mathcal{M})italic_μ ∉ italic_S ( caligraphic_M ). Then, we have three cases to analyze:

    1. (a)

      If μ(w)=φi𝜇𝑤subscript𝜑𝑖\mu(w)=\varphi_{i}italic_μ ( italic_w ) = italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT, then there is jJi𝑗subscript𝐽𝑖j\in J_{i}italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT such that w=λ(fij)𝑤𝜆subscript𝑓𝑖𝑗w=\lambda(f_{ij})italic_w = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ). Assume that Pwφisubscript𝑃𝑤subscript𝜑𝑖\emptyset P_{w}\varphi_{i}∅ italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT, then by the construction of the one-to-one associated market (particularly condition (1) on workers’ preferences) it follows that P¯wfijsubscript¯𝑃𝑤subscript𝑓𝑖𝑗\emptyset\overline{P}_{w}f_{ij}∅ over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, contradicting that λS(M)𝜆superscript𝑆𝑀\lambda\in S^{*}(M)italic_λ ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ).

    2. (b)

      Let wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ). Assume that wCφi(μ(φi))𝑤subscript𝐶subscript𝜑𝑖𝜇subscript𝜑𝑖w\notin C_{\varphi_{i}}(\mu(\varphi_{i}))italic_w ∉ italic_C start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ). Since μ(φi)=jλ(fij)𝜇subscript𝜑𝑖subscript𝑗𝜆subscript𝑓𝑖𝑗\mu(\varphi_{i})=\bigcup_{j}\lambda(f_{ij})italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) = ⋃ start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) and Cφ(μ(φi))=jmaxμ(φ)Pijsubscript𝐶𝜑𝜇subscript𝜑𝑖subscript𝑗subscript𝜇𝜑subscript𝑃𝑖𝑗C_{\varphi}(\mu(\varphi_{i}))=\bigcup_{j}\max_{\mu(\varphi)}P_{ij}italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ) = ⋃ start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT, then there is jJsuperscript𝑗𝐽j^{\prime}\in Jitalic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J such that λ(fij)=w𝜆subscript𝑓𝑖superscript𝑗𝑤\lambda(f_{ij^{\prime}})=witalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) = italic_w and maxμ(φi)Pijwsubscript𝜇subscript𝜑𝑖subscript𝑃𝑖superscript𝑗𝑤\max_{\mu(\varphi_{i})}P_{ij^{\prime}}\neq wroman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ≠ italic_w. Then, λ𝜆\lambdaitalic_λ is blocked by fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT since it satisfies condition (2) of Definition 1, contradicting that λS(M)𝜆superscript𝑆𝑀\lambda\in S^{*}(M)italic_λ ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ).

    3. (c)

      Assume that there is a blocking pair to μ𝜇\muitalic_μ, i.e., there is a pair (φi,w)subscript𝜑𝑖𝑤(\varphi_{i},w)( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_w ) such that φiPwμ(w)=φisubscript𝜑𝑖subscript𝑃𝑤𝜇𝑤subscript𝜑superscript𝑖\varphi_{i}P_{w}\mu(w)=\varphi_{i^{\prime}}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_μ ( italic_w ) = italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT for some isuperscript𝑖i^{\prime}italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT, and wCφ(μ(φi){w}).𝑤subscript𝐶𝜑𝜇subscript𝜑𝑖𝑤w\in C_{\varphi}(\mu(\varphi_{i})\cup\{w\}).italic_w ∈ italic_C start_POSTSUBSCRIPT italic_φ end_POSTSUBSCRIPT ( italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ∪ { italic_w } ) . Then, by the Aizermann-Malishevski decomposition, there is ij𝑖superscript𝑗ij^{\prime}italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT such that w=maxμ(φi){w}Pij𝑤subscript𝜇subscript𝜑𝑖𝑤subscript𝑃𝑖superscript𝑗w=\max_{\mu(\varphi_{i})\cup\{w\}}P_{ij^{\prime}}italic_w = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ∪ { italic_w } end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT which implies that wPijλ(fij′′)𝑤subscript𝑃𝑖superscript𝑗𝜆subscript𝑓𝑖superscript𝑗′′wP_{ij^{\prime}}\lambda(f_{ij^{\prime\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each j′′Ji{j}superscript𝑗′′subscript𝐽𝑖superscript𝑗j^{\prime\prime}\in J_{i}\setminus\{j^{\prime}\}italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ∖ { italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT }. By the construction of the associated one-to-one market, if φiPwφisubscript𝜑𝑖subscript𝑃𝑤subscript𝜑superscript𝑖\varphi_{i}P_{w}\varphi_{i^{\prime}}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT, then fijP¯wfij~subscript𝑓𝑖superscript𝑗subscript¯𝑃𝑤subscript𝑓superscript𝑖~𝑗f_{ij^{\prime}}\overline{P}_{w}f_{i^{\prime}\tilde{j}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT over~ start_ARG italic_j end_ARG end_POSTSUBSCRIPT for each j~Ji~𝑗subscript𝐽superscript𝑖\tilde{j}\in J_{i^{\prime}}over~ start_ARG italic_j end_ARG ∈ italic_J start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT (in particular for λ(w)=fij^𝜆𝑤subscript𝑓superscript𝑖^𝑗\lambda(w)=f_{i^{\prime}\hat{j}}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT over^ start_ARG italic_j end_ARG end_POSTSUBSCRIPT with j^Ji^𝑗subscript𝐽superscript𝑖\hat{j}\in J_{i^{\prime}}over^ start_ARG italic_j end_ARG ∈ italic_J start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT). Hence, (w,fij)𝑤subscript𝑓𝑖superscript𝑗(w,f_{ij^{\prime}})( italic_w , italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) is a blocking* pair to λ𝜆\lambdaitalic_λ, contradicting that λS(M)𝜆superscript𝑆𝑀\lambda\in S^{*}(M)italic_λ ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ).

  2. (ii)

    Given μS(),thenT𝟏(μ)=λS(M)formulae-sequence𝜇𝑆𝑡𝑒𝑛superscript𝑇1𝜇𝜆superscript𝑆bold-∗𝑀\boldsymbol{\mu\in S\left(\mathcal{M}\right)$,then$T^{-1}(\mu)=\lambda\in S^{% \ast}\left(M\right)}bold_italic_μ bold_∈ bold_italic_S bold_( bold_caligraphic_M bold_) bold_, bold_italic_t bold_italic_h bold_italic_e bold_italic_n bold_italic_T start_POSTSUPERSCRIPT bold_- bold_1 end_POSTSUPERSCRIPT bold_( bold_italic_μ bold_) bold_= bold_italic_λ bold_∈ bold_italic_S start_POSTSUPERSCRIPT bold_∗ end_POSTSUPERSCRIPT bold_( bold_italic_M bold_). We prove this implication by proving two more claims.

    Claim 3

    λ𝜆\lambdaitalic_λ is a one-to-one matching.

    By way of contradiction, assume λ𝜆\lambdaitalic_λ is not a matching. So, there is w𝒲𝑤𝒲w\in\mathcal{W}italic_w ∈ caligraphic_W such that λ(w)=fij𝜆𝑤subscript𝑓𝑖𝑗\lambda(w)=f_{ij}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT and λ(w)=fij𝜆𝑤subscript𝑓superscript𝑖superscript𝑗\lambda(w)=f_{i^{\prime}j^{\prime}}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. If ii𝑖superscript𝑖i\neq i^{\prime}italic_i ≠ italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT, then wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) and wμ(φi)𝑤𝜇subscript𝜑superscript𝑖w\in\mu(\varphi_{i^{\prime}})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ), which would contradict that μ𝜇\muitalic_μ is a many-to-one matching. If i=i𝑖superscript𝑖i=i^{\prime}italic_i = italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT and jj𝑗superscript𝑗j\neq j^{\prime}italic_j ≠ italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT, then either j<j𝑗superscript𝑗j<j^{\prime}italic_j < italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT or vice-versa, making λ𝜆\lambdaitalic_λ contradict the construction of T1(μ)superscript𝑇1𝜇T^{-1}(\mu)italic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_μ ).

    The construction of T1superscript𝑇1T^{-1}italic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ensures the bilateral nature of the assignment.

    Claim 4

    λS(M).𝜆superscript𝑆𝑀\lambda\in S^{*}(M).italic_λ ∈ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ) .

    By way of contradiction, assume λS(M)𝜆superscript𝑆𝑀\lambda\notin S^{*}(M)italic_λ ∉ italic_S start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ( italic_M ).

    1. (a)

      Assume that λ𝜆\lambdaitalic_λ is blocked* by a worker w𝑤witalic_w. If λ(w)=fij𝜆𝑤subscript𝑓𝑖𝑗\lambda(w)=f_{ij}italic_λ ( italic_w ) = italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT then wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ). Since μS()𝜇𝑆\mu\in S(\mathcal{M})italic_μ ∈ italic_S ( caligraphic_M ), then φiPwsubscript𝜑𝑖subscript𝑃𝑤\varphi_{i}P_{w}\emptysetitalic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∅. By construction of the preferences in the associated one-to-one market, it follows that fijP¯wsubscript𝑓𝑖𝑗subscript¯𝑃𝑤f_{ij}\overline{P}_{w}\emptysetitalic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT ∅ for each jJi𝑗subscript𝐽𝑖j\in J_{i}italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT, a contradiction.

    2. (b)

      Assume that λ𝜆\lambdaitalic_λ is blocked* by a firm-copy. Since μS()𝜇𝑆\mu\in S(\mathcal{M})italic_μ ∈ italic_S ( caligraphic_M ) then Cφi(μ(φi))=μ(φi)=jJimaxμ(φi)Pijsubscript𝐶subscript𝜑𝑖𝜇subscript𝜑𝑖𝜇subscript𝜑𝑖subscript𝑗subscript𝐽𝑖subscript𝜇subscript𝜑𝑖subscript𝑃𝑖𝑗C_{\varphi_{i}}(\mu(\varphi_{i}))=\mu(\varphi_{i})=\bigcup_{j\in J_{i}}\max_{% \mu(\varphi_{i})}P_{ij}italic_C start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ) = italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) = ⋃ start_POSTSUBSCRIPT italic_j ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT. By construction of T1superscript𝑇1T^{-1}italic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT if λ(fij)=w𝜆subscript𝑓𝑖superscript𝑗𝑤\lambda(f_{ij^{\prime}})=witalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) = italic_w, then maxμ(φi)Pij=wPijλ(fij′′)subscript𝜇subscript𝜑𝑖subscript𝑃𝑖superscript𝑗𝑤subscript𝑃𝑖superscript𝑗𝜆subscript𝑓𝑖superscript𝑗′′\max_{\mu(\varphi_{i})}P_{ij^{\prime}}=wP_{ij^{\prime}}\lambda(f_{ij^{\prime% \prime}})roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT = italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each j′′Ji{j}superscript𝑗′′subscript𝐽𝑖superscript𝑗j^{\prime\prime}\in J_{i}\setminus\{j^{\prime}\}italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ∖ { italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT } and wPij𝑤subscript𝑃𝑖superscript𝑗wP_{ij^{\prime}}\emptysetitalic_w italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ∅, a contradiction.

    3. (c)

      Assume that there are fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT and fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT such that λ(fij)Pijλ(fij)Pij.𝜆subscript𝑓𝑖superscript𝑗subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗\lambda(f_{ij^{\prime}})P_{ij}\lambda(f_{ij})P_{ij}\emptyset.italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅ . Since μS()𝜇𝑆\mu\in S(\mathcal{M})italic_μ ∈ italic_S ( caligraphic_M ), and T1(μ)=λsuperscript𝑇1𝜇𝜆T^{-1}(\mu)=\lambdaitalic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_μ ) = italic_λ, we have that for each wμ(φi)𝑤𝜇subscript𝜑𝑖w\in\mu(\varphi_{i})italic_w ∈ italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) there is Pijsubscript𝑃𝑖𝑗P_{ij}italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT such that w=maxμ(φi)Pij𝑤subscript𝜇subscript𝜑𝑖subscript𝑃𝑖𝑗w=\max_{\mu(\varphi_{i})}P_{ij}italic_w = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT and w=λ(fij).𝑤𝜆subscript𝑓𝑖𝑗w=\lambda(f_{ij}).italic_w = italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) . Then, for each fijsubscript𝑓𝑖𝑗f_{ij}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT such that λ(fij)Pij𝜆subscript𝑓𝑖𝑗subscript𝑃𝑖𝑗\lambda(f_{ij})P_{ij}\emptysetitalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ∅ there is no fijsubscript𝑓𝑖superscript𝑗f_{ij^{\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT such that λ(fij)Pijλ(fij)𝜆subscript𝑓𝑖superscript𝑗subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖𝑗\lambda(f_{ij^{\prime}})P_{ij}\lambda(f_{ij})italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ), a contradiction.

    4. (d)

      Assume that (w,fij)𝑤subscript𝑓𝑖𝑗(w,f_{ij})( italic_w , italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) form a blocking* pair, i.e., wPijλ(fij)𝑤subscript𝑃𝑖𝑗𝜆subscript𝑓𝑖superscript𝑗wP_{ij}\lambda(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJisuperscript𝑗subscript𝐽𝑖j^{\prime}\in J_{i}italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT and fijP¯wλ(w)subscript𝑓𝑖𝑗subscript¯𝑃𝑤𝜆𝑤f_{ij}\overline{P}_{w}\lambda(w)italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_λ ( italic_w ).444 W.l.o.g. assume j=min{j′′Jij=\min\{j^{\prime\prime}\in J_{i}italic_j = roman_min { italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT such that wPij′′λ(fij)𝑤subscript𝑃𝑖superscript𝑗′′𝜆subscript𝑓𝑖superscript𝑗wP_{ij^{\prime\prime}}\lambda(f_{ij^{\prime}})italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT ) for each jJi}.j^{\prime}\in J_{i}\}.italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT } . Also assume that some firm-copy is matched, otherwise stability and stability* coincide. By construction of T1superscript𝑇1T^{-1}italic_T start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT if λ(fij)𝜆subscript𝑓𝑖𝑗\lambda(f_{ij})\neq\emptysetitalic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) ≠ ∅ then λ(fij)=maxμ(φi)Pij𝜆subscript𝑓𝑖𝑗subscript𝜇subscript𝜑𝑖subscript𝑃𝑖𝑗\lambda(f_{ij})=\max_{\mu(\varphi_{i})}P_{ij}italic_λ ( italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ) = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT. It follows that wPij[maxμ(φi)Pij]𝑤subscript𝑃𝑖𝑗delimited-[]subscript𝜇subscript𝜑𝑖subscript𝑃𝑖𝑗wP_{ij}\left[\max_{\mu(\varphi_{i})}P_{ij}\right]italic_w italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT [ roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT ], which in turn implies that w=maxμ(φi){w}Pij𝑤subscript𝜇subscript𝜑𝑖𝑤subscript𝑃𝑖𝑗w=\max_{\mu(\varphi_{i})\cup\{w\}}P_{ij}italic_w = roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ∪ { italic_w } end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT. Then, we have that wCφi(μ(φi){w})=jJimaxμ(φi){w}Pij𝑤subscript𝐶subscript𝜑𝑖𝜇subscript𝜑𝑖𝑤subscriptsuperscript𝑗subscript𝐽𝑖subscript𝜇subscript𝜑𝑖𝑤subscript𝑃𝑖superscript𝑗w\in C_{\varphi_{i}}(\mu(\varphi_{i})\cup\{w\})=\bigcup_{j^{\prime}\in J_{i}}% \max_{\mu(\varphi_{i})\cup\{w\}}P_{ij^{\prime}}italic_w ∈ italic_C start_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ∪ { italic_w } ) = ⋃ start_POSTSUBSCRIPT italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ∈ italic_J start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT end_POSTSUBSCRIPT roman_max start_POSTSUBSCRIPT italic_μ ( italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) ∪ { italic_w } end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_i italic_j start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. By construction of the associated one-to-one market, fijP¯wfij′′subscript𝑓𝑖𝑗subscript¯𝑃𝑤subscript𝑓superscript𝑖superscript𝑗′′f_{ij}\overline{P}_{w}f_{i^{\prime}j^{\prime\prime}}italic_f start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_f start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT italic_j start_POSTSUPERSCRIPT ′ ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT implies φiPwφisubscript𝜑𝑖subscript𝑃𝑤subscript𝜑superscript𝑖\varphi_{i}P_{w}\varphi_{i^{\prime}}italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT italic_P start_POSTSUBSCRIPT italic_w end_POSTSUBSCRIPT italic_φ start_POSTSUBSCRIPT italic_i start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. Then, (w,φi)𝑤subscript𝜑𝑖(w,\varphi_{i})( italic_w , italic_φ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) is a blocking pair to μ𝜇\muitalic_μ, contradicting that μS()𝜇𝑆\mu\in S(\mathcal{M})italic_μ ∈ italic_S ( caligraphic_M ).

By (i) and (ii), T𝑇Titalic_T is an isomorphism between the set of stable many-to-one matchings and the set of stable* one-to-one matchings.   \square

The following example illustrates the isomorphism between the two markets of Example 1.

Example 1 (continued) Recall that the stable matchings of the many-to-one subsitutable market are:

μΦ=(φ1φ2w1w2w3w4)μ1=(φ1φ2w1w3w2w4)subscript𝜇Φmatrixsubscript𝜑1subscript𝜑2subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4subscript𝜇1matrixsubscript𝜑1subscript𝜑2subscript𝑤1subscript𝑤3subscript𝑤2subscript𝑤4\displaystyle\mu_{\Phi}=\begin{pmatrix}\varphi_{1}&\varphi_{2}\\ w_{1}w_{2}&w_{3}w_{4}\end{pmatrix}\mu_{1}=\begin{pmatrix}\varphi_{1}&\varphi_{% 2}\\ w_{1}w_{3}&w_{2}w_{4}\end{pmatrix}italic_μ start_POSTSUBSCRIPT roman_Φ end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG ) italic_μ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG )
μ2=(φ1φ2w2w4w1w3)μ𝒲=(φ1φ2w3w4w1w2)subscript𝜇2matrixsubscript𝜑1subscript𝜑2subscript𝑤2subscript𝑤4subscript𝑤1subscript𝑤3subscript𝜇𝒲matrixsubscript𝜑1subscript𝜑2subscript𝑤3subscript𝑤4subscript𝑤1subscript𝑤2\displaystyle\mu_{2}=\begin{pmatrix}\varphi_{1}&\varphi_{2}\\ w_{2}w_{4}&w_{1}w_{3}\end{pmatrix}\mu_{\mathcal{W}}=\begin{pmatrix}\varphi_{1}% &\varphi_{2}\\ w_{3}w_{4}&w_{1}w_{2}\end{pmatrix}italic_μ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG ) italic_μ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_φ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL italic_φ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG )

The stable* matchings of the associated one-to-one market are:

λ=(f11f12f13f14f15f16f21f22f23f24f25f26w1w2w3w4)subscript𝜆matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\lambda_{\mathcal{F}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{1}&\emptyset&\emptyset&w_{2}&\emptyset&\emptyset&w_{3}&\emptyset&\emptyset&% w_{4}&\emptyset&\emptyset\end{pmatrix}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG )
λ1=(f11f12f13f14f15f16f21f22f23f24f25f26w2w4w3w1)subscript𝜆1matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤2subscript𝑤4subscript𝑤3subscript𝑤1\lambda_{1}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{12}&f_{13}&% f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{2}&\emptyset&w_{4}&\emptyset&\emptyset&\emptyset&w_{3}&\emptyset&\emptyset&% \emptyset&\emptyset&w_{1}\end{pmatrix}italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL end_ROW end_ARG )
λ2=(f11f12f13f14f15f16f21f22f23f24f25f26w1w3w4w2)subscript𝜆2matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤1subscript𝑤3subscript𝑤4subscript𝑤2\lambda_{2}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{12}&f_{13}&% f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{1}&\emptyset&\emptyset&\emptyset&\emptyset&w_{3}&w_{4}&\emptyset&w_{2}&% \emptyset&\emptyset&\emptyset\end{pmatrix}italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG )
λ𝒲=(f11f12f13f14f15f16f21f22f23f24f25f26w3w4w2w1)subscript𝜆𝒲matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤3subscript𝑤4subscript𝑤2subscript𝑤1\lambda_{\mathcal{W}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{3}&w_{4}&\emptyset&\emptyset&\emptyset&\emptyset&w_{2}&w_{1}&\emptyset&% \emptyset&\emptyset&\emptyset\end{pmatrix}italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG )

It easy to see that T(λ)=μΦ𝑇subscript𝜆subscript𝜇ΦT(\lambda_{\mathcal{F}})=\mu_{\Phi}italic_T ( italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ) = italic_μ start_POSTSUBSCRIPT roman_Φ end_POSTSUBSCRIPT, T(λ1)=μ1𝑇subscript𝜆1subscript𝜇1T(\lambda_{1})=\mu_{1}italic_T ( italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT ) = italic_μ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT, T(λ2)=μ2𝑇subscript𝜆2subscript𝜇2T(\lambda_{2})=\mu_{2}italic_T ( italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ) = italic_μ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT, and T(λ𝒲)=μ𝒲.𝑇subscript𝜆𝒲subscript𝜇𝒲T(\lambda_{\mathcal{W}})=\mu_{\mathcal{W}}.italic_T ( italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ) = italic_μ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT . \Diamond

4 Final remarks

In this paper, we present a method to decompose a many-to-one market with substitutable preferences into a one-to-one market using the Aizerman-Malishevski decomposition. We define a notion of stability* for an associated one-to-one market that is adjusted to show that the set of stable matchings in a many-to-one market and the set of stable* matchings in the associated one-to-one market are isomorphic. Furthermore, we present an adaptation of the deferred acceptance algorithm (originally introduced by Gale and Shapley, 1962) for our associated one-to-one market. Regardless of which side proposes –be it the firm-copies or the workers– a stable* matching is always obtained. Although the fact that both markets are isomorphic already indicates that the set of stable* matchings in the related one-to-one market is non-empty, the algorithm provides an alternative proof of this fact.

A common result when using the deferred acceptance algorithm across all matching markets is that the output is proposing-side optimal: if, for instance, the firms are the ones making the offers, the output of the algorithm is the firm-optimal stable matching. Unfortunately, this is not valid in our associated one-to-one market. Recall from Example 1 the stable* matchings resulting from the adapted deferred acceptance algorithm when both the firm-copies and the workers propose:

λ=(f11f12f13f14f15f16f21f22f23f24f25f26w1w2w3w4)subscript𝜆matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤1subscript𝑤2subscript𝑤3subscript𝑤4\lambda_{\mathcal{F}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{1}&\emptyset&\emptyset&w_{2}&\emptyset&\emptyset&w_{3}&\emptyset&\emptyset&% w_{4}&\emptyset&\emptyset\end{pmatrix}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG )

and

λ𝒲=(f11f12f13f14f15f16f21f22f23f24f25f26w3w4w2w1),subscript𝜆𝒲matrixsubscript𝑓11subscript𝑓12subscript𝑓13subscript𝑓14subscript𝑓15subscript𝑓16subscript𝑓21subscript𝑓22subscript𝑓23subscript𝑓24subscript𝑓25subscript𝑓26subscript𝑤3subscript𝑤4subscript𝑤2subscript𝑤1\lambda_{\mathcal{W}}=\setcounter{MaxMatrixCols}{12}\begin{pmatrix}f_{11}&f_{1% 2}&f_{13}&f_{14}&f_{15}&f_{16}&f_{21}&f_{22}&f_{23}&f_{24}&f_{25}&f_{26}\\ w_{3}&w_{4}&\emptyset&\emptyset&\emptyset&\emptyset&w_{2}&w_{1}&\emptyset&% \emptyset&\emptyset&\emptyset\end{pmatrix},italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = ( start_ARG start_ROW start_CELL italic_f start_POSTSUBSCRIPT 11 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 13 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 14 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 15 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 16 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 21 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 22 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 23 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 24 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 25 end_POSTSUBSCRIPT end_CELL start_CELL italic_f start_POSTSUBSCRIPT 26 end_POSTSUBSCRIPT end_CELL end_ROW start_ROW start_CELL italic_w start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL italic_w start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_CELL start_CELL italic_w start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL start_CELL ∅ end_CELL end_ROW end_ARG ) ,

respectively. If we observe, for instance, the firm f12subscript𝑓12f_{12}italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT, we find that

w4=λ𝒲(f12)P12λ(f12)=,subscript𝑤4subscript𝜆𝒲subscript𝑓12subscript𝑃12subscript𝜆subscript𝑓12w_{4}=\lambda_{\mathcal{W}}(f_{12})P_{12}\lambda_{\mathcal{F}}(f_{12})=\emptyset,italic_w start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT = italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT ) italic_P start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ( italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT ) = ∅ , (2)

indicating that the optimality of λsubscript𝜆\lambda_{\mathcal{F}}italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT fails.

Let us consider the following result on isomorphic lattices: Given two partially ordered sets (A,>A)𝐴subscript𝐴(A,>_{A})( italic_A , > start_POSTSUBSCRIPT italic_A end_POSTSUBSCRIPT ) and (B,>B)𝐵subscript𝐵(B,>_{B})( italic_B , > start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT ) and an isomorphism \mathcal{I}caligraphic_I between the sets A𝐴Aitalic_A and B𝐵Bitalic_B that preserves the order: for a,bA𝑎𝑏𝐴a,b\in Aitalic_a , italic_b ∈ italic_A such that a>Absubscript𝐴𝑎𝑏a>_{A}bitalic_a > start_POSTSUBSCRIPT italic_A end_POSTSUBSCRIPT italic_b, it follows that (a)>B(b)subscript𝐵𝑎𝑏\mathcal{I}(a)>_{B}\mathcal{I}(b)caligraphic_I ( italic_a ) > start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT caligraphic_I ( italic_b ), if (A,>A)𝐴subscript𝐴(A,>_{A})( italic_A , > start_POSTSUBSCRIPT italic_A end_POSTSUBSCRIPT ) has a lattice structure, then (B,>B)𝐵subscript𝐵(B,>_{B})( italic_B , > start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT ) also has a lattice structure (see Birkhoff, 1967, for a thorough treatment of lattice theory).

If we consider the orders P𝒲subscript𝑃𝒲P_{\mathcal{W}}italic_P start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT and P¯𝒲subscript¯𝑃𝒲\overline{P}_{\mathcal{W}}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT induced by workers’ preferences in the many-to-one and related one-to-one markets, we can assert that the set of stable* matchings has a lattice structure with respect to the induced order P¯𝒲subscript¯𝑃𝒲\overline{P}_{\mathcal{W}}over¯ start_ARG italic_P end_ARG start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT since the ordering between matchings is preserved. However, the previous example also shows that the same cannot be stated for the firm-copies side. Despite that μΦsubscript𝜇Φ\mu_{\Phi}italic_μ start_POSTSUBSCRIPT roman_Φ end_POSTSUBSCRIPT is unanimously preferred (either considering partial order induced by firms’ preferences or considering Blair’s partial order (Blair, 1988)) by all firms to μ𝒲subscript𝜇𝒲\mu_{\mathcal{W}}italic_μ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT in the many-to-one market, μΦ=T(λ)subscript𝜇Φ𝑇subscript𝜆\mu_{\Phi}=T(\lambda_{\mathcal{F}})italic_μ start_POSTSUBSCRIPT roman_Φ end_POSTSUBSCRIPT = italic_T ( italic_λ start_POSTSUBSCRIPT caligraphic_F end_POSTSUBSCRIPT ), and μ𝒲=T(λ𝒲)subscript𝜇𝒲𝑇subscript𝜆𝒲\mu_{\mathcal{W}}=T(\lambda_{\mathcal{W}})italic_μ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT = italic_T ( italic_λ start_POSTSUBSCRIPT caligraphic_W end_POSTSUBSCRIPT ), (2) shows that the isomorphisms T𝑇Titalic_T do not preserve the order.

Several important conclusions from this paper may guide future research. One significant avenue is the establishment of an order among matchings on the firm-copies side, which we think has to be closely related to Blair’s partial order (see Blair, 1988, for more details). This new possible order is essential for demonstrating that the set of many-to-many matchings with substitutable preferences forms a lattice structure. Once a new order is established for matchings in the related one-to-one market on the firm-copies side, we can ask for firms preferences in the original many-to-one model not only be substitutable but also satisfy the well-known “law of aggregated demand” (LAD).555The law of aggregated demand states that when a firm chooses from an expanded set, it selects at least as many workers as before. This property was first studied by Alkan (2002) under the name ’cardinal monotonicity’. See also Hatfield and Milgrom (2005). This requirement could open numerous avenues for future research. Below, we outline some of these potential directions:

  1. (i)

    It may allow for the definition of pointing functions for the associated one-to-one market, which would enable the calculation of the join and meet between two stable* matchings (see Martínez et al., 2001; Alkan, 2002; Li, 2014, for more details on pointing functions in many-to-many markets).

  2. (ii)

    By defining pointing functions in the associated one-to-one market, it may be possible to provide an appropriate definition of side-optimality, ensuring that the output of the adapted deferred acceptance algorithm is side-optimal.

  3. (iii)

    Bonifacio et al. (2022) presents an algorithm that computes, using the notion of cycles in preferences, the entire set of stable matchings for a many-to-many market with substitutable preferences that satisfy the LAD (this algorithm is a generalization of the one presented in Irving and Leather, 1986). This algorithm runs the lattice of the set of stable matchings, from the firm-optimal to the worker-optimal stable matching. We believe that by requiring the LAD on firms’ preferences (in the many-to-many market), it may be possible to identify a cycle structure in the preferences of firm-copies (in the associated one-to-one market), thus it may be possible to develop an algorithm that computes the entire set of stable* matchings.

  4. (iv)

    Another important result for matching markets is the well-known “Rural Hospitals Theorem”.666The Rural Hospital Theorem is proven in different contexts by many authors (see McVitie and Wilson, 1971; Roth, 1984, 1985; Martínez et al., 2000; Alkan, 2002, among others). The version of this theorem for a many-to-many matching market, where all agents have substitutable choice functions satisfying the LAD, which also applies to our setting, is presented in Alkan (2002). It states that each agent is matched with the same number of partners in every stable matching. It is coherent that this theorem does not hold in the associated one-to-one market since, in the many-to-one market, without requiring the LAD on firms’ preferences beyond substitutability, the result does not hold. However, as we can observe in Example 1 (the firms’ preferences in the many-to-one market satisfy both substitutability and LAD), although this result does not hold for a firm-copy (see the case of the firm-copy f12subscript𝑓12f_{12}italic_f start_POSTSUBSCRIPT 12 end_POSTSUBSCRIPT in (2)), we can observe that the number of matched copies of a firm is the same in every stable* matching. Therefore, it may be possible to adapt this result to the associated one-to-one market when requiring both substitutability and the LAD in the preferences of firms in the original many-to-one market.

References

  • Aizerman and Malishevski (1981) Aizerman, M. and A. Malishevski (1981): “General theory of best variants choice: Some aspects,” IEEE Transactions on Automatic Control, 26, 1030–1040.
  • Alkan (2002) Alkan, A. (2002): “A class of multipartner matching markets with a strong lattice structure,” Economic Theory, 19, 737–746.
  • Birkhoff (1967) Birkhoff, G. (1967): Lattice theory, American Mathematical Society.
  • Blair (1988) Blair, C. (1988): “The Lattice Structure of the Set of Stable Matchings with Multiple Partners,” Mathematics of Operations Research, 13, 619–628.
  • Bonifacio et al. (2022) Bonifacio, A. G., N. Juarez, P. Neme, and J. Oviedo (2022): “Cycles to compute the full set of many-to-many stable matchings,” Mathematical Social Sciences, 117, 20–29.
  • Chambers and Yenmez (2017) Chambers, C. P. and M. B. Yenmez (2017): “Choice and matching,” American Economic Journal: Microeconomics, 9, 126–147.
  • Crawford and Kelso (1982) Crawford, V. P. and A. Kelso (1982): “Job Matching, Coalition Formation, and Gross Substitutes,” Econometrica, 50, 1483–1504.
  • Gale and Shapley (1962) Gale, D. and L. Shapley (1962): “College admissions and the stability of marriage,” American Mathematical Monthly, 69, 9–15.
  • Hatfield and Milgrom (2005) Hatfield, J. W. and P. R. Milgrom (2005): “Matching with Contracts,” American Economic Review, 95, 913–935.
  • Irving and Leather (1986) Irving, R. and P. Leather (1986): “The complexity of counting stable marriages,” SIAM Journal of Computing, 15, 655–667.
  • Li (2014) Li, J.-R. (2014): “A new proof of the lattice structure of many-to-many pairwise-stable matchings,” Journal of the Operations Research Society of China, 2, 369–377.
  • Martínez et al. (2000) Martínez, R., J. Massó, A. Neme, and J. Oviedo (2000): “Single Agents and the Set of Many-to-One Stable Matchings,” Journal of Economic Theory, 91, 91–105.
  • Martínez et al. (2001) ——— (2001): “On the lattice structure of the set of stable matchings for a many-to-one model,” Optimization, 50, 439–457.
  • McVitie and Wilson (1971) McVitie, D. G. and L. B. Wilson (1971): “Three procedures for the stable marriage problem,” Communications of the ACM, 14, 491–492.
  • Moulin (1985) Moulin, H. (1985): “Choice functions over a finite set: A summary,” Social Choice and Welfare, 2, 147–160.
  • Roth (1984) Roth, A. E. (1984): “The Evolution of the Labor Market for Medical Interns and Residents: A Case Study in Game Theory,” Journal of Political Economy, 92, 991.
  • Roth (1985) ——— (1985): “The college admissions problem is not equivalent to the marriage problem,” Journal of Economic Theory, 36, 277–288.
  • Roth and Sotomayor (1990) Roth, A. E. and M. Sotomayor (1990): Two-sided matching, Cambridge University Press.