Do we need a welfare state at all?
Those with a preference for minimal state intervention are likely to suggest a combination of two options: private insurance, and charity to take care of those in need of assistance. Those two options share a number of advantages: both types of providers can be more local, and they are competing with others of their kind, which, over time, should make them much more efficient than a giant state welfare bureaucracy ever could. Both increase people’s freedom to choose how they would like to use their financial resources – whether, and if so, how much, to pay to which insurance company, and deciding the exact monetary value one assigns to the well-being of others in society, respectively.
While this may sound great to free-market enthusiasts, each of these suggestions brings its own problems. Let us begin with private insurance. The catchphrase here is incomplete information: it leads to both adverse selection and moral hazard. The former would make private insurance provision almost useless, as only those with very high risks will demand insurance at the high premia companies would be forced to charge, rendering the business unprofitable. The second concerns the behavioral change that individuals may exhibit once insured.
In this respect, it is considerably harder to make a private insurance market work in welfare than in, say, car insurance. An insured driver may be less careful than an uninsured one, but they would not crash their own car on purpose. But some people may actually value being unemployed while being paid, for example as a form of early retirement. Barr (1992) cites another, related problem: a share of potential insurance clients will have a probability of unemployment close to unity, so that no private company is willing to provide insurance to these individuals. These problems can be prevented through making insurance mandatory both on the demand side and on the supply side. In practice, forcing private companies to take on high-risk, unprofitable clients necessitates transfer payments to these companies. One may consider this an option preferable to a full supply of unemployment insurance by the state, and indeed, there are good examples of healthcare systems that work exactly like this (namely the social health insurance systems of the Netherlands, Switzerland, Germany, Belgium and Israel). However, while these are market-based models, they are not the completely free insurance market that libertarian purists would have in mind.
The second option of non-state welfare provision is charity, for example organised through the church. This essentially suggests reviving a model of welfare provision which was predominant centuries ago, before the rise of the modern welfare state. As touched on before, this model enables individuals to decide how much they personally value the well-being of those in need. The rationale behind this is that individuals’ factor in the utility of other into their own utility function, so that a gift from one person to the other can make both better off in utility terms. However, as Milton Friedman (cited in Barr, 1992, see footnote) explains, if the overall well-being of society is important to those who give, private charity cannot be optimal. The reason is that others will freeride on those who give, by seeing a better-off society without having contributed to it. This in turn demotivates those giving in the first place, because they do not want to be the only ones contributing and the results remain behind what they could be if everyone contributed. Welfare, then, has properties of a public good.
Of course, the point can be made that this system has, in fact, functioned to a satisfactory degree in the past and could therefore do so again. Changing social patterns like the declining role of church institutions aside, this reasoning can potentially have a very anti-libertarian effect: it is likely to tie people to their local charity providers, thereby discouraging them from moving for the chance of employment. Such a decrease in labor mobility can hardly be seen as increasing people’s freedom or helping the efficient allocation of labor resources. The fragmentation of the system, while desirable in other respects, also risks inefficient allocation, by enabling “benefit shopping” – individuals receiving overlapping benefits from several charities, which are unaware of this due to the lack of an overarching registration system (this happens, for example, with local microfinance providers in developing countries).
Finally, let us come back to the lesson learned from the “restaurant analogy” in the first part of this blog series: a predictable benefit system, where the effect of getting a job or increasing working hours can be foreseen accurately, is essential to increasing employment. With a diversity of providers, such predictability is lacking: getting a low-paid job might mean losing all benefits from one provider, while another may just reduce benefits. The lack of predictability and thus created search costs of finding the most generous provider may deter recipients from taking up employment. A critical reader will now object that search costs could be lower for most buyer choices if the state dictates supply, yet surely that is the last thing most of us want? Naturally, this is correct, and this marks the fundamental difference between unemployment insurance and most other products and services: a consumer in a market will weigh search costs, mobility costs and others against the benefits of the transaction and, if they are excessive, may either not participate in the transaction or not switch providers once they have settled for one. This is not usually a circumstance that society has to be overly concerned with. In the case of unemployment insurance however, if such costs prevent benefit recipients from entering employment, this could cause vast inefficiencies. It remains to weigh the obvious inefficiencies of the current, public provision of unemployment benefits against the inefficiencies of its various private alternatives as mentioned here.
In pursuit of striking a balance, it seems that Universal Credit has made a move in the right direction in several aspects of efficiency enhancement. However, there remains room for improvement in order to bring as many people back into employment as possible, and debates fought in other countries can be an inspiration here.
Carolin Bollig is a research intern at the IEA.
 Barr, N. (1992). Economic Theory and the Welfare State: A Survey and Interpretation. Journal of Economic Literature, 30 (2).