Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE ( (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 25 October 2020

Public Finance and Soft Budgets

Abstract and Keywords

The concept of soft budget constraint, describes a situation where a decision-maker finds it impossible to keep an agent to a fixed budget. In healthcare it may refer to a (nonprofit) hospital that overspends, or to a lower government level that does not balance its accounts. The existence of a soft budget constraint may represent an optimal policy from the regulator point of view only in specific settings. In general, its presence may allow for strategic behavior that changes considerably its nature and its desirability. In this article, soft budget constraint will be analyzed along two lines: from a market perspective and from a fiscal federalism perspective.

The creation of an internal market for healthcare has made hospitals with different objectives and constraints compete together. The literature does not agree on the effects of competition on healthcare or on which type of organizations should compete. Public hospitals are often seen as less efficient providers, but they are also intrinsically motivated and/or altruistic. Competition for quality in a market where costs are sunk and competitors have asymmetric objectives may produce regulatory failures; for this reason, it might be optimal to implement soft budget constraint rules to public hospitals even at the risk of perverse effects. Several authors have attempted to estimate the presence of soft budget constraint, showing that they derive from different strategic behaviors and lead to quite different outcomes.

The reforms that have reshaped public healthcare systems across Europe have often been accompanied by a process of devolution; in some countries it has often been accompanied by widespread soft budget constraint policies. Medicaid expenditure in the United States is becoming a serious concern for the Federal Government and the evidence from other states is not reassuring. Several explanations have been proposed: (a) local governments may use spillovers to induce neighbors to pay for their local public goods; (b) size matters: if the local authority is sufficiently big, the center will bail it out; equalization grants and fiscal competition may be responsible for the rise of soft budget constraint policies. Soft budget policies may also derive from strategic agreements among lower tiers, or as a consequence of fiscal imbalances. In this context the optimal use of soft budget constraint as a policy instrument may not be desirable.

Keywords: soft budget constraint, healthcare, fiscal imbalance, decentralization, hospital competition, public hospitals, patients mobility, health economics


The concept of soft budget constraint, which was firstly formulated by Kornai (1986) to describe economic behavior in socialist economies, is nowadays applied to describe any situation where a decision-maker finds it impossible to keep an agent to a fixed budget.1

Borrowed from the terminology of microeconomics, it suggests the presence of an agency relationship where the agent interprets its budget as soft; in other words, it does not take its budget limit too seriously. The reasons leading to this behavior are heterogeneous: in some cases, the uncertainty in some key factors (revenue or costs) makes it impossible for the agent to respect its budget constraints, while in others the nature of the good produced may induce the regulator to bail out the inefficient provider. Finally, it might be the principal itself that promotes such behavior in order to allow more flexibility to the provision of a specific service. The theory of the soft budget constraint has been applied to several contexts at both micro and macro levels. This article will focus on healthcare, a sector where, due to its peculiar characteristics, soft budget constraints practices arise at both micro and macroeconomic levels.

In healthcare, the term “soft budget constraint” may refer to several situations, for example a (nonprofit) hospital that overspends or a lower government level that does not balance its accounts. Kornai (2009) and Crivelli et al. (2010) provide interesting reviews of the problem and of the main actors involved. In the short run, soft budget constraint practices may allow reduction of the negative effects of budget cuts and of the uncertainty that characterizes the provision of healthcare. However, their use, even when it is advocated on welfare improvement grounds, leads to strategic behaviors that in the long run may have perverse effects on the organization of healthcare.

In what follows, the soft budget constraint will be analyzed along two lines: from a market perspective, involving the decision of a single hospital to run into a deficit as well as the strategy of a regulator to bail out public hospitals, and from a fiscal federalism perspective, involving a lower government tier that decides to run into a deficit.

The Market Perspective

In order to understand the role of soft budgets in a market perspective, it is necessary to recall the reasons leading to the creation of an internal market for hospital care and the peculiarities that prevent market forces from working in this context. Internal markets have been created for publicly provided healthcare in an attempt to shift power from producers to consumers in a context where consumers have very weak incentives to seek out low-cost producers (healthcare is free at the point of use, or is supplied in exchange for a copayment covering only a fraction of the cost of care) and have little knowledge about the quality of healthcare (they are infrequent users and key information is withheld by producers on both professional and commercial grounds). The objective of these reforms was to reduce the inefficiencies of vertical integration and to improve quality through competition. In actual fact, these market have proved to be much less competitive than expected, mostly because providers are quite different both in their objectives and in their constraints.

The literature does not agree on the effects of competition on healthcare (Enthoven, 2002; Gaynor et al., 2015; Gaynor & Town, 2011; Goddard, 2015; Siciliani, 2018) or on which type of organizations should compete (Levaggi, 2007; Brekke et al., 2011; Herr, 2011; Brekke et al., 2012; Levaggi & Levaggi, 2017b). This uncertainty has made the organization of the market rather country-specific, the main differences being the nature of the bodies created by the reform, their independence in decision-making, the contract rules for the provision of services, and the regulation of competition with private suppliers. Whichever their form, true competitive markets cannot be created for the well-known reasons presented above.

An interesting feature of internal markets relates to heterogeneity in the providers, which have different objectives and different constraints. Public hospitals are often seen as less efficient providers, but they may be altruistic (Brekke et al., 2011) or intrinsically motivated (Delfgaauw, 2007; Galizzi et al., 2015). Competition for quality in a market where costs are sunk and competitors have asymmetric objectives produces regulatory failures, and it might be optimal to implement soft budget constraint rules to public hospitals (Brekke et al., 2015; Levaggi & Levaggi, 2017a). It should be stressed that this policy may be optimal only in a context where hospitals are not able to foresee bailing out. If this is not the case, soft budget rules decrease total welfare. For this reason, in a repeated game where all the actors can play strategically, the desired effects of soft budget constraint may be sensibly reduced.

First, the models showing that soft budget may be welfare-improving will be examined. The environment in which competition is set is characterized by the following assumptions:

  • healthcare is free at the point of use (or patients pay a small fraction of its cost);

  • competition among providers is made on quality, the latter being a variable that can be observed by the patient but that cannot be verified by the purchaser;

  • providers are reimbursed using DRG-type payments which incorporate the cost for medical equipment (Levaggi et al., 2014);

  • providers do not share the same objectives;

  • Hotelling/Salop models of spatial competition are used.

The asymmetry in objectives and constraints is a crucial point that determines the results in terms of competition and efficiency. A growing literature (Delfgaauw, 2007; Galizzi et al., 2015) suggests that the objectives of public and private providers may be quite different. While in the past this difference was mostly modeled to capture issues of surplus versus quantity (Herr, 2011), more recent approaches argue that the difference may depend on the different motivations of the staff. In particular, public providers embed intrinsic motivations in their function. Intrinsic motivation in healthcare means that providers share with patients some objectives, usually measured in terms of quality. Workers are more motivated and willing to work for a lower salary (Barigozzi & Burani, 2016; Delfgaauw & Dur, 2010); this may not necessarily lead to lower costs for several reasons (Brekke et al., 2011, 2012; Delfgaauw & Dur, 2008; Levaggi et al., 2009), but changes their objective function and their way of competing. Public hospitals usually cannot retain all the surplus they make; their motivation to reduce cost is greatly reduced, but quality enhancement is an intrinsic objective, not a means to get more market share.

Brekke et al. (2015) propose a model where hospitals may be profit-maximizing, or they may have asymmetric objectives; costs are uncertain, and patients are heterogeneous in their costs (i.e., a fixed proportion of patients have a higher severity/cost, the rest a lower level). Hospitals compete for patients in a Hotelling framework and the proportion of patients with high costs is not known beforehand.

The authors first consider a Nash equilibrium where the hospitals earn negative profits in the high-demand state and positive profits in the low-demand state. Soft budget is introduced by assuming that the government (sponsor), with some probability, will bail out the hospital ex post if it runs a deficit. Softer budgets, by reducing the expected deficit in the high-demand state, weaken hospitals’ incentives for cost-efficiency, but they may improve hospital quality. On the other hand, profit confiscation is always bad for hospital quality and cost-efficiency. In this setting, the authors show that allowing public hospitals to run a deficit (soft budget constraint) enhances the quality they provide; as a result, private hospitals (which have an hard budget constraint) will increase quality as well. This policy increases costs, but it is welfare-improving. This result is reinforced in a setting where public hospitals are intrinsically motivated.

Levaggi and Levaggi (2017a) propose a different explanation which relies on the presence of fixed investment in medical technology whose cost is reimbursed through a DRG payment scheme (Levaggi et al., 2014). The competition framework in which the model is set is a Salop competition with a public hospital at the center. The cost to provide care is made of two parts: a fixed investment H and a variable cost that may be reduced through efficiency improvements. In the benchmark solution, only a public hospital provides the services and the DRG payment is set so as to balance the hospital accounts. Let us now assume that a mixed market, where hospitals compete for patients, is created. Let us also assume that the reimbursement T is fixed to the benchmark level. For private hospitals, H is an entry cost: if profits are sufficiently high to cover it, they will enter the market.

For public hospitals, this is not an option; they have no incentive to reduce the running costs because they cannot retain any profit, hence they will run into a deficit because T is no longer sufficient to make them break even. Depending on the relative efficiency of the two providers, the second-best solution for this market is to make public and private hospitals compete and let the public hospital run into a deficit; this is in fact the solution that maximizes quality and reduces costs. The alternative solution would to increase T to make the public hospital break even, but the profit of private hospitals would increase as well.

These conclusions are valid only if hospitals do not foresee that the regulator will bail them out. When this assumption is ruled out, deficits may increase and the risk of perverse effects is very high, as shown below.

Levaggi and Montefiori (2013) consider a mixed market for hospital care where public providers coexist with private ones. Private hospitals maximize the surplus, while the public hospital’s objective is represented by the maximization of several alternative objectives: surplus (as with private hospitals), market share, and “equity” defined in terms of the quality offered to different types of patients under different setting concerning the budget constraint of public hospitals. In this model, patients are heterogeneous in the severity of their ailment (low and high severity). Hospitals provide two types of quality: hotel and medical quality. High-severity patients have a bias for medical quality, while low-severity patients prefer hotel quality. In this framework it is possible to show that if public hospitals do not pursue surplus maximization and the regulator allows them to run a deficit, hospitals have an incentive in selecting their case mix in equilibrium. In general, private hospitals choose a higher level of hotel quality in order to attract low-cost patients, while public hospitals increase medical quality and attract high cost patients. As a result, private hospitals increase their surplus at the expenses of public hospitals, which have to be bailed out by the regulator.

The models reviewed so far are set in a static, timeless framework where it is reasonable to assume that hospitals may not foresee bailouts. In actuality, contracts with providers have a long-term nature where contractual obligations are set over several periods. In this framework, strategic behavior and anticipation of the opponent’s move are quite likely.

This case is examined by Wright (2016) in a context where a public hospital supplies healthcare. The regulator sets the budget, while the hospital sets the number of patients to be treated. In the presence of a hard budget constraint, the regulator sets the budget for a specific time period T and the hospital reacts with a set number of patients. If the regulator is able to set a welfare-maximizing budget, the response of the hospital will also maximize welfare.

If the budget constraint is soft, the public hospital sets the number of patients and the time t in which it will exhaust it. If tT, the game ends. If instead the public hospital decides to exhaust the budget before T, a second-period game is set where the regulator decides to give the hospital extra money, but this process has costs in terms of extra resources that should be devoted to healthcare. According to Wright, the size of the bailout costs is the key variable in determining whether the hospital will over-treat. However, if this is the case, government may be able to anticipate this behavior and set an initial budget below or above the welfare-maximizing level; in this case, the number of patients will be either “too low” or “too high” in the first period.

In an interesting development of the model, the author considers that the costs of bailout are a function of the length of the bailout period itself. In this case, by optimally choosing the budget for the first period the regulator may reduce the perverse effects of the soft budget constraint. Finally, when private hospitals enter the picture as second-choice destinations for the patients that the public hospital does not treat, the size of the bailout is reduced and welfare is increased.

To sum up, at market level bailing out seems to arise because of the asymmetries in the behavior of the different players, a problem that may have been overlooked at the time of the reforms that introduced competition in hospital care and to which there might not be a clear solution. Healthcare is an essential good; any policy threat of not bailing out over-spenders may not be credible, because it would undermine healthcare provision.

However, this is not the only source of bailing out in the provision of healthcare: the organization of healthcare often requires delegated choices over different government levels, and this may create other soft budget issues.

Fiscal Federalism

The reforms that have reshaped public healthcare systems across Europe have often been accompanied by a process of reallocation of functions among different government tiers. In some countries devolution has often been characterized by widespread soft budget constraint policies (Rodden et al., 2003; Besfamille & Lockwood, 2008). Medicaid expenditure in the United States is becoming a serious concern for the Federal Government (Wildasin & Marton, 2007) and the evidence from other states is not reassuring (Finck & Stratmann, 2009; Sorribas-Navarro, 2011; Crivelli et al., 2010).

In a decentralized context, there are two main levels at which bailout may occur. One is across jurisdictions: local governments use spillovers to induce neighbors to pay for their local public goods (Wildasin, 2001, 2004). In other words, some local authorities free ride on the provision of essential services. The quantity produced will be suboptimal, but the local authority is better off in the short run. The other is in the relationship between different levels of governments. In this case, lower government tiers induce a higher one to pay for their expenses, whether because size matters (Crivelli & Staal, 2013), because of fiscal imbalances (Costa-Font, 2013), for political accountability problems (Arends, 2017), or for a combination of these reasons. In these cases, bailouts are necessary because central governments cannot credibly allow subnational governments to go bankrupt. Hard budget constraint rules would in fact prevent the provision of an essential service such as healthcare. In this context, the central government should reduce the probability of bailout by using other incentives to make lower tiers more responsible for their expenditure decisions.

In healthcare, the presence of soft budget constraint policies depends on an organizational failure. Traditional literature on fiscal federalism (Oates, 2008) suggests that decentralization should follow efficiency principles. Expenditure decisions should be left to the tier which is better informed on local preferences, while grants might be used for equity and efficiency reasons. Furthermore, fiscal federalism should induce some inter-jurisdictional competition among political powers resulting from “vote with the feet” or yardstick competition. Second-generation models (see Oates, 2005 for a review) suggest that the success of fiscal federalism depends on the information the agents possess about either specific parameters, the behavior of other agents, or the effects of their decisions on total welfare. Most of the literature on fiscal federalism assumes that the good to be produced is a local public good with spillovers.

However, healthcare is an impure public goods with spillovers. Impure public goods are both private goods (increasing utility for the quantity actually bought) and public goods (for the entire amount produced); the latter are private goods whose consumption is financed by the government for equity/redistribution purposes. These goods can be made available to a specific community either by producing them or by allowing people to receive them outside the local authority boundaries. Fiscal federalism in this context has particular characteristics that have not received due attention. Firstly, by its nature, the equalization grant and political accountability will play a very important role, especially in contexts where income is unevenly distributed. Secondly, cross border provision gives rise to financial agreements that need to be regulated and that may themselves be the cause for the soft budget constraint (Brekke et al., 2014, 2016).

The problems arising in the devolution of healthcare are well explained in Petretto (2000), which compares the welfare properties of a unitary system where central government set total expenditure and local provision and a regional system where each region can decide its level of local production.

The first interesting conclusion of this analysis is that the desirability of regionalization from the region’s point of view may not coincide with the central government’s view nor with its desirability from the overall social point of view.

From the regional government’s point of view, the desirability of regionalization usually derives from a comparison between the benefits and the costs of public funds. In the benefits formula, the revenue from the regional “export” of healthcare services to other regions (the payment for the spillover effects) as well as the self-financing effect of regionalization should also be considered. The first element depends on the price set for regional mobility, which in this context should also take into account the spillover created by healthcare. The second element depends on the rules that define the equalization grant and on the regional disparity in income. This is the reason why the desirability of regionalization from the overall social point of view is strongly influenced by the considerations of redistribution aims that should be included in the evaluation of social benefits. This is the basic problem of decentralization in the presence of impure public goods with spillover and regional income inequality. The lack of correspondence between regional and national objectives may lead to claims for devolution that are not justified on efficiency reasons (Levaggi & Menoncin, 2017) or to strategic behavior leading to soft budget constraint behavior.

Let us first consider bailout as a strategic game among jurisdictions, as proposed by Wildasin (2004). Healthcare is a local public good, but its benefits spread over to the other jurisdiction, or the citizens in one region moves to the other to get the benefits of that good. Let us also assume that the price for healthcare is equal to one so that Qi represents both expenditure and quantity. The welfare for each local authority can be written as:


In general, each local authority should buy at least the level of expenditure that maximizes its welfare. The optimal quantity is found for the level for which the marginal utility is equal to marginal cost; that is, fiQi=1. For healthcare, we can assume that f and g are quite similar, that is, the utility of the good produced in the other region is very similar to the one derived from utility of the quantity locally produced. Let us also assume that f1<f2. In this case, local authority 1 may play strategically. By knowing that the other local authority cares about health, it will reduce the optimal quantity supplied while the other will increase it. In this way, Region 1 maximizes its welfare at a cost of a reduction in the welfare of the entire community.

Crivelli and Staal (2013) propose a model that start from similar assumptions, but considers the relationship between different government tiers. The model assumes the presence of N regions that are different in their size. There are economies of scale in the provision of local public goods so that some regions may prefer zero supply (the quantity produced is not sufficient to make the price competitive) and rely on the spillovers of the other regions. Local provision is always suboptimal, because local authorities take into account only local benefits, and the zero quantity may arise for this reason. If spillovers are sufficiently high, the central government may decide to subsidize local provision using a matching grant which, by reducing the price paid for the local public good, increases local provision. However, if local authorities foresee this policy, they may strategically reduce the quantity of local public good produced in order to receive a higher matching grant. The authors show that this policy depends on the size of the local authority and on the spillovers. First of all, the local authority should be sufficiently big to use the economies of scale, otherwise the central government would not choose to finance local provision in the first place. Secondly, the spillovers should be relatively high to justify the extra budget that the central government has to set aside.

On a similar line, Bordignon and Turati (2009) develop a model where local government expectations on being bailed out are relevant. In this case there are two level of government (central and local), two level of expenditure that can be selected by the local authority (low and high), and two level of grants from the center to the lower tier (low and high). If the center chooses high, the local authority also chooses high. The strategic game arises when the grant is low and expenditure is high because a deficit arises. In a full information system, this event would never occur; on the other hand, when asymmetry of information exists and when the no-bailout policy is credible (for economic or political reasons), if the local government is sufficiently sure of being bailed out it will run a deficit by increasing expenditure even when the grant is not sufficient to pay for all provisions.

Costa-Font (2013) argues that soft budget constraint arises from vertical fiscal imbalances. Subnational governments often decide the level of expenditure on local services, but may not have enough resources to provide them because of an uneven distribution of income at the local level. If this is the case, the central government provides additional resources using grants. However, the grant is not a neutral instrument. The author suggests that rules are often not transparent; what follows presents a slightly different explanation that relates to the type of grant used for healthcare.

The most popular instrument used for equalization is the expenditure-based grant (Blöchliger & Charbit, 2008; Vaillancourt & Bird, 2010; Levaggi & Menoncin, 2017), which for a two-regions setting can be written as:


where τi is the tax rate and Yi the income of each region. Let us now assume that the welfare of each region is a linear combination of net income and the utility of healthcare (h), which is modeled as a local public good with spillovers. In order to set expenditure for healthcare, Region i maximizes its welfare subject to its budget constraint; the problem can be written as:




Bailing out depends on the ability of each region to foresee the effects that its own expenditure will have on the equalization grant (ρ). Thus, 1ρ can be considered as a measure of fiscal illusion or strategic play. Expenditure on local public good will then be equal to2


When the region does not play strategically, the demand is equal to what we would expect from a Cobb Douglas; that is, it reflects local preferences for the public good. However, if the local authority plays strategically, the demand will also reflect its contribution to the equalization grant. The richer region (Yi>Yj) reduces the optimal provision, while the less rich region increases it knowing that the equalization grant received will be higher; in other words, it will induce the central government to bail it out. This may explain why some countries, such as Switzerland, have to use matching grants to induce rich regions to produce healthcare, while at the same time some other regions may overspend.

Fiscal imbalances may also cause political accountability problems. When the equalization grant is the most important part of local government finance, the use of local resources to finance healthcare is discouraged, because the cost (in political and economic term) of increasing the tax rate has very little effect on expenditure—for the local authority it makes more sense to increase expenditure and the deficit. In this case, it is in fact likely that the center will bail the region out; in fact, the only alternative for the upper tier is to centralize the function—the region is so poor that any threat to make it pay for its own debits is not credible.

Horizontal and vertical bailouts are combined in the model presented by Levaggi and Menoncin (2012). Richer (and more efficient) regions and poorer (and less efficient) ones play strategically to force the central government to bail regional deficits out. The former wish to control overall expenditure and the equalization grant, but they have capacity in excess of local needs; the latter wish to minimize the effort in raising resources locally and in becoming more efficient. Healthcare mobility from poor to rich regions can be used to achieve both goals. Patient mobility is used as the main variable to induce the central government to bail out the regions that run into a deficit. The less efficient local authorities prefer to send their citizens to receive services outside their region. This policy is supported by the more efficient region, which, due to the shape of its utility function, prefers to produce more goods than are locally needed. However, patient mobility is financed out of deficit instead of local resources; this means that unless the central government bails out with extra resources the less efficient regions, the more efficient one will never be paid for the nonresident patients treated. In this case, both less efficient and more efficient regions have an interest in lobbying for a bailout. In the short run, richer local authorities can control the growth of healthcare expenditure and do not have to reduce their excess capacity, while poor local authorities can shift a part of their tax bill to the central government. However, the lack of coordination between local objectives and total welfare means that this policy is optimal at local level, but it produces an overall welfare loss. There are in fact two clear losers: the whole community, which would be better off if hard budget constraint rules were imposed, and the users of the services in the regions where soft budget constraint is widespread, who have to travel and incur private costs.

Empirical Evidence

Several models have been proposed to test for the presence of the soft budget constraint, and empirical analyses usually study single-country evidence. On the market side, Hagen and Kaarboe (2006) show that in Norway supplementary funding was common in the years preceding and following the hospital reforms of 2002. The centralization of ownership, financing, and production should have increased budgetary discipline; however, Tjerbo and Hagen (2009) show that this result has not been accomplished: production has been far above what was planned, and the deficits have become higher than ever.

Kornai (2009) shows that in Hungary about 30% of hospitals were in financial trouble in 2007, but through time the government has become less and less open to bailing out hospitals. The strategy they usually follow is exhaust their budgets before the end of the budget period or run up debts by the end of the budget period, knowing that the government will provide additional funding. Eggleston and Shen (2011) show that soft budgets are not a characteristic of public hospitals, but the presence of a higher number of public hospital reduces the probability that private hospitals run into a deficit. The authors also show that hardening the constraint may have perverse effects on quality of care, especially for the poor and the frail. In fact, private hospitals, in the quest to budget their balance, may start to reduce the supply of services that are less remunerative and to introduce cream-skimming/dumping policies. Soft budget constraint may also alter the managerial structure of the hospitals, and they may spur different behavior according to their status. Eldenburg et al. (2017) show that hospitals’ responses to budget issues not only depend on its objectives (nonprofit vs. for-profit hospitals), but also on diversity within a specific group (religious, secular, nonprofit university status), an evidence that may open new lines of research into this field.

The empirical evidence for soft budget at regional level are rather scant, since they mainly depend on the architecture that central government has chosen for healthcare (centralization vs. devolution). Bordignon and Turati (2009) present empirical evidence for the soft budget constraint at the regional level. They argue that public health expenditure in Italy is (partly) the result of a strategic game being played by regional and central governments. Local governments form expectations on the probability of being bailed out, which depend on the presence of external constraints. The model is tested for Italian data and the authors study the impact of a number of specific regional variables on the formation of bailout expectations. The results show that richer and more autonomous regions had lower expectations of central government intervention. Interestingly, regions which had the same political majority of the central government in charge reduced health expenditure more than those run by “unfriendly” ones.

Soft budget policies may also derive from strategic agreements among lower tiers, as shown in Levaggi and Menoncin (2012). Their model is tested with Italian data for the period 2002–2006, and the results confirm the intuition of the theoretical model. From 2007 in Italy, the central government has tightened its budget rules. The regions that have a deficit have to present a plan (Piano di Rientro) to reduce their deficits; if they do not comply, the central government may directly increase its tax rate to meet expenditure. The empirical evidence on the effects of this policy is still scant, but most regions (especially those that were close to balancing their budgets) have been able to curb their expenditure, and they are trying to reduce mobility through bilateral agreements with the other regions, setting a limit to the number of patients that are treated outside the region.

For Spain, Costa-Font (2013) shows that bailout is strictly connected with fiscal imbalances, which in turn makes a hard budget constraint threat less credible. Like Italy, Spain is characterized by a significant variance in income at the regional level, and in general this seems the leitmotif of the soft budget in decentralized healthcare provision.


Public healthcare expenditure has been steadily growing as a percentage of GDP, and the process has been accompanied by devolution of functions at both the micro level (with the creation of internal markets for hospital care) and the macro level (with more functions allocated to lower government tiers). This process has given rise to coordination problems which may cause soft budget constraints. The expectation of bailouts weakens budget constraints and induces agents to behave strategically when selecting healthcare expenditure, quality, and other key variables. In general, they cause a loss in welfare, and they may considerably change the desired outcomes of government intervention in the market for healthcare.

This article has examined the decision to spend beyond the budget limits in two different contexts by examining the reasons that inform the decision of a hospital to overspend and by studying the behavior of subnational governments.

At the hospital level, soft budget constraint policies do not necessarily produce a welfare loss; in fact, this may be the best solution to cope with uncertainty and the asymmetry in objectives and constraints that characterize the market for healthcare. However, when being bailed out becomes a variable that is taken into account by the hospitals in setting their activity level, the beneficial effects of soft budget constraints are always lost.

The key issue for the success of this policy is the ability of hospitals to foresee bailouts. If this is the case, soft budget constraint may have perverse effects and should be avoided. Government may find it difficult not to grant bailouts to public hospitals, especially when they are big and there is no oversupply in the area where they are located. In this case, the hospital cannot go bankrupt; their closure would restrict access to hospital care. However, other actions may be possible, from sacking the managers to publishing statistics on the efficiency level of each hospital. These instruments may act as moral suasions in reducing the undesired effect of soft budget constraints.

From a fiscal federalism point of view, soft budget constraints are more likely when a high share of subnational spending is financed from the common pool of federal resources and in the presence of vertical fiscal imbalances. In this case, welfare is always suboptimal, but enforcing hard budget constraint rules, especially in the short run, may be rather challenging for a central government. The nature of the good produced makes it difficult for the center to be credible in denying bailout to jurisdictions that run into a deficit, and political accountability may prevent it from imposing a hard budget. In order to reduce strategic behavior, it may be advisable to reduce the use of expenditure-based grants in favor of resource-based ones. Germany and Switzerland have recently reformed the equalization base for healthcare in this direction, and it will be interesting to see the effects on expenditure and bailouts.

Another important factor in this game is determined by voters’ behavior. In some countries, although the choices as concern healthcare have been delegated to local governments, citizens may still feel that health is a national good and they may think that the central government is ultimately responsible for its provision. In this case, their vote at the national level would be influenced by the local provision, giving even more strength to local decision-makers in their bailout requests. In the presence of fiscal imbalances, the use of fiscal federalism may not be advisable, and the choice of some governments, such as that of the United Kingdom, to choose a more centralized architecture may allow them to keep expenditure under control.


The author would like to thank the reviewers for their helpful comments and the copyeditor Stephen Dodson. They have contributed to greatly improve the paper. The usual disclaimer applies.


Arends, H. (2017). More with less? Fiscal decentralisation, public health spending and health sector performance. Swiss Political Science Review, 23(2), 144–174.Find this resource:

    Barigozzi, F., & Burani, N. (2016). Competition and screening with motivated health professionals. Journal of Health Economics, 50, 358–371.Find this resource:

      Besfamille, M., & Lockwood, B. (2008). Bailouts in federations: Is a hard budget constraint always best? International Economic Review, 49(2), 577–593.Find this resource:

        Blöchliger, H., & Charbit, C. (2008). Fiscal equalisation. OECD Economic Studies 44, OECD.Find this resource:

          Bordignon, M., & Turati, G. (2009). Bailing out expectations and public health expenditure. Journal of Health Economics, 28(2), 305–321.Find this resource:

            Brekke, K. R., Levaggi, R., Siciliani, L., & Straume, O. R. (2014). Patient mobility, health care quality and welfare. Journal of Economic Behavior & Organization, 105, 140–157.Find this resource:

              Brekke, K. R., Levaggi, R., Siciliani, L., & Straume, O. R. (2016). Patient mobility and health care quality when regions and patients differ in income. Journal of Health Economics, 50, 372–387.Find this resource:

                Brekke, K. R., Siciliani, L., & Straume, O. R. (2011). Quality competition with profit constraints: Do non-profit firms provide higher quality than for-profit firms? Technical Report 8284, C.E.P.R. Discussion Papers.Find this resource:

                  Brekke, K. R., Siciliani, L., & Straume, O. R. (2012). Quality competition with profit constraints. Journal of Economic Behavior & Organization, 84(2), 642–659.Find this resource:

                    Brekke, K. R., Siciliani, L., & Straume, O. R. (2015). Hospital competition with soft budgets. Scandinavian Journal of Economics, 117(3), 1019–1048.Find this resource:

                      Breuillé, M.-L., & Vigneault, M. (2010). Overlapping soft budget constraints. Journal of Urban Economics, 67(3), 259–269.Find this resource:

                        Costa-Font, J. (2013). Decentralization and the Spanish health system: Soft budget constraint modernization? London, UK: Palgrave Macmillan UK.Find this resource:

                          Crivelli, E., Leive, A., & Stratmann, T. (2010). Subnational health spending and soft budget constraints in OECD countries. IMF Working Papers 10/147, International Monetary Fund.Find this resource:

                            Crivelli, E., & Staal, K. (2013). Size, spillovers and soft budget constraints. International Tax and Public Finance, 20(2), 338–356.Find this resource:

                              Delfgaauw, J. (2007). Dedicated doctors: Public and private provision of health care with altruistic physicians. Tinbergen Institute Discussion Papers 07-010/1, Tinbergen Institute.Find this resource:

                                Delfgaauw, J., & Dur, R. (2008). Incentives and workers motivation in the public sector. The Economic Journal, 118(525), 171–191.Find this resource:

                                  Delfgaauw, J., & Dur, R. (2010). Managerial talent, motivation, and self-selection into public management. Journal of Public Economics, 94(9–10), 654–660.Find this resource:

                                    Eggleston, K., & Shen, Y.-C. (2011). Soft budget constraints and ownership: Empirical evidence from US hospitals. Economics Letters, 110(1), 7–11.Find this resource:

                                      Eldenburg, L., Krishnan, H. A., & Krishnan, R. (2017). Management accounting and control in the hospital industry: A review. Journal of Governmental & Nonprofit Accounting, 6(1), 52–91.Find this resource:

                                        Enthoven, A. (2002). Introducing market forces into health care: A tale of two countries. Paper delivered to the Fourth European Conference on Health Economics, Paris.Find this resource:

                                          Finck, A., & Stratmann, T. (2009). Institutionalized bailouts and fiscal policy: The consequences of soft budget constraints. CESifo Working Paper Series 2827, CESifo Group Munich.Find this resource:

                                            Galizzi, M. M., Tammi, T., Godager, G., Linnosmaa, I., & Wiesen, D. (2015). Provider altruism in health economics. THL Discussion Paper 4, National Institute for Health and Welfare, Helsinki, Finland.Find this resource:

                                              Gaynor, M., Ho, K., & Town, R. J. (2015). The industrial organization of health-care markets. Journal of Economic Literature, 53(2), 235–284.Find this resource:

                                                Gaynor, M., & Town, R. J. (2011). Competition in health care markets. In M. V. Pauly, T. G. McGuire, & P. P. Barros (Eds.), Handbook of health economics (Vol. 2, pp. 499–637). Amsterdam, The Netherlands: Elsevier.Find this resource:

                                                  Goddard, M. (2015). Competition in healthcare: Good, bad or ugly? International Journal of Health Policy and Management, 4(9), 567–569.Find this resource:

                                                    Hagen, T. P., & Kaarboe, O. M. (2006). The Norwegian hospital reform of 2002: Central government takes over ownership of public hospitals. Health Policy, 76(3), 320–333.Find this resource:

                                                      Herr, A. (2011). Quality and welfare in a mixed duopoly with regulated prices: The case of a public and a private hospital. German Economic Review, 12(4), 422–437.Find this resource:

                                                        Kornai, J. (1986). The soft budget constraint. Kyklos, 39(1), 3–30.Find this resource:

                                                          Kornai, J. (2009). The soft budget constraint syndrome in the hospital sector. International Journal of Health Care Finance and Economics, 9(2), 117–135.Find this resource:

                                                            Kornai, J., Maskin, E., & Roland, G. (2003). Understanding the soft budget constraint. Journal of Economic Literature, 41(4), 1095–1136.Find this resource:

                                                              Levaggi, L. & Levaggi, R. (2017a). Oligopolistic competition for the provision of hospital care. Technical report, SSRN working paper.Find this resource:

                                                                Levaggi, L., & Levaggi, R. (2017b). Regulation strategies for the provision of paternalistic goods. Annals of Public and Cooperative Economics, 88(4), 471–498.Find this resource:

                                                                  Levaggi, R. (2007). Regulating internal markets for hospital care. Journal of Regulatory Economics, 32(2), 173–193.Find this resource:

                                                                    Levaggi, R., & Menoncin, F. (2012). Soft budget constraints in health care: Evidence from Italy. The European Journal of Health Economics, 14(5), 725–737.Find this resource:

                                                                      Levaggi, R., & Menoncin, F. (2017). Would less regional income distribution justify the present call for devolution? International Tax and Public Finance, 24(5), 780–799.Find this resource:

                                                                        Levaggi, R., & Montefiori, M. (2013). Patient selection in a mixed oligopoly market for health care: The role of the soft budget constraint. International Review of Economics, 60(1), 49–70.Find this resource:

                                                                          Levaggi, R., Moretto, M., & Pertile, P. (2014). Two part payments for the reimbursement of investments in health technologies. Health Policy, 115(2–3), 230–236.Find this resource:

                                                                            Levaggi, R., Moretto, M., & Rebba, V. (2009). Investment decisions in hospital technology when physicians are devoted workers. Economics of Innovation and New Technology, 18(5), 487–512.Find this resource:

                                                                              Oates, W. E. (2005). Toward a second-generation theory of fiscal federalism. International Tax and Public Finance, 12, 349–373.Find this resource:

                                                                                Oates, W. E. (2008). On the evolution of fiscal federalism: Theory and institutions. National Tax Journal, 61, 313–334.Find this resource:

                                                                                  Petretto, A. (2000). On the cost-benefit of the regionalisation of the National Health Service. Economics of Governance, 1(3), 213–232.Find this resource:

                                                                                    Pettersson-Lidbom, P. (2010). Dynamic commitment and the soft budget constraint: An empirical test. American Economic Journal: Economic Policy, 2(3), 154–179.Find this resource:

                                                                                      Rodden, J. A., Eskeland, G. S., & Litvack, J. (2003). Introduction and overview. In J. A. Rodden & G. S. Eskeland (Eds.), Fiscal decentralization and the challenge of hard budget constraints (pp. 3–32). Cambridge, MA: MIT Press.Find this resource:

                                                                                        Siciliani, L. (2018). Hospital economics: Competition, tariffs and non-profit status. In B. H. Baltagi & F. Moscone (Eds.), Health econometrics (forthcoming). Bingley, UK: Emerald Publishing.Find this resource:

                                                                                          Sorribas-Navarro, P. (2011). Bailouts in a fiscal federal system: Evidence from Spain. European Journal of Political Economy, 27(1), 154–170.Find this resource:

                                                                                            Tjerbo, T., & Hagen, T. P. (2009). Deficits, soft budget constraints and bailouts: Budgeting after the Norwegian hospital reform. Scandinavian Political Studies, 32(3), 337–358.Find this resource:

                                                                                              Vaillancourt, F., & Bird, R. (2010). Expenditure based equalization transfers. In J. Martinez-Vazquez & B. Searle (Eds.), Fiscal equalization: Challenges in the design of intergovernmental transfers (2nd ed., pp. 259–289). Boston, MA: Springer.Find this resource:

                                                                                                Wildasin, D., & Marton, J. (2007). Medicaid expenditures and state budgets: Past, present, and future. Working Papers 2007–04, University of Kentucky, Institute for Federalism and Intergovernmental Relations.Find this resource:

                                                                                                  Wildasin, D. E. (2001). Externalities and bailouts: Hard and soft budget constraints in intergovernmental fiscal relations. Public Economics 0112002, EconWPA.Find this resource:

                                                                                                    Wildasin, D. E. (2004). The institutions of federalism: Towards an analytical framework. National Tax Journal, 57, 247–272.Find this resource:

                                                                                                      Wright, D. J. (2016). Soft budget constraints in public hospitals. Health Economics, 25(5), 578–590.Find this resource:


                                                                                                        (1.) See Kornai et al. (2003) for an extensive review of the concept and its application.

                                                                                                        (2.) The complete model is presented in Levaggi and Menoncin (2017).