By David A. Robalino, Lead Economist — Labor and Youth Team Leader, the World Bank
As classic unemployment insurance (UI) spreads in the developing world, the debate is heating up over the best way to design unemployment benefit programs. It’s an issue that I’ve been working on for some time, but unfortunately, there has been little good evidence about the impact of these programs on workers' incentives and their ability to find jobs. Most of the research so far has referred to OECD countries, where UI is the most common public income support program for the unemployed. But the situation is more complicated in developing countries, with their large informal sectors, which offer a relatively easy way for unemployed people to pick up some income — undetected by the government — while they continue to receive UI benefits.
Construction Workers in Brazil. June 16, 2009 © World Bank Photo Collection
A recent paper
on UI in Argentina turns up some badly needed evidence and offers an interesting policy recommendation. The authors, Hernan Ruffo and Martin Gonzalex-Rozada (Torcuato Di Tella University), examine how UI — funded in Argentina by a 1.5 percent payroll tax on employers — affects unemployment duration and reemployment wages. It shows, as expected by many, that the generosity of the unemployment benefit (given by the level and duration) increases the duration of the unemployment spell, but it doesn't really help workers find better jobs (such as those that pay a higher wage). However, the level of the benefit does seem to improve wages somewhat, although not the unemployment duration. These findings suggest, as Ruffo recently told the JKP, that policy makers should err on the side of offering more generous benefits instead of longer durations.
But I think we can do even better in solving the fundamental problem with classic UI — the existence of both implicit subsidies to workers who don't do their best in trying to find jobs and implicit taxes on the "savings" of workers who spend less time unemployed.
Relying on a mix of insurance and savings accounts
Another way to tackle the problem, which Michael Weber and I pursue in a recent paper
, is to make the subsidies and taxes explicit, and then try to find a combination that provides sufficient protection to workers and better incentives to find jobs. This can be done by simply tracking what each individual contributes to and gets out of the system; basically calculating the balance in their "unemployment accounts." In the traditional UI (like in Argentina), the positive balances are taxed at a 100 percent rate to cover the accounts in deficit. We suggest reducing the tax on savings and instead using general revenues or consumption taxes.
Some of you might recognize that this is similar to the unemployment individual savings accounts (UISAs)
, a more recent instrument, which is typically found in Latin America countries like Chile and Brazil. In UISAs the tax on savings is zero but workers can't have negative balances in their accounts — that is, they receive unemployment benefits as long as they have savings. The problem is that for certain workers it isn't easy to accumulate sufficient savings to finance adequate unemployment benefits. Thus, a given level of subsidies is needed (which Chile does through a solidarity fund).
However, the model we propose is a more general design that has classic UI and classic UISAs as two extremes — in one the tax on savings is zero, in the other it is 100 percent. But there is no reason to fall into either of these extremes. What countries need to decide is the optimal level of benefits and their duration (the Argentinean paper provides some guidance) and then find the best way to finance the subsidies needed to cover those with negative balances — a tax on savings can be part of the financing mechanism.
In addition, conditionalities to receive benefits should be limited to the participation in job-search and training activities — and not include employment status, which is very difficult to enforce when there are large informal sectors. New technologies such as smart cards with biometric identification can be used to monitor these conditionalities.
Hopefully, this way of think about unemployment benefits will facilitate the, sometimes ideological, debate between the virtues and flaws of UI and UISAs.