Note ¹

This is an important paper on an important topic. The author convincingly argues, on the basis of both theory and careful empirical work, that Poland’s minimum wage scheme has contributed to that country’s high rate of unemployment and has hindered an otherwise successful process of transition. I agree with this conclusion. The lessons are important for Poland, and they are important for all countries in transition, present or future. Cuba, in particular, will not be able to afford labor market policies that keep unemployment high at a time when wage flexibility will be needed to smooth the transition and when budgetary resources will be scarce. Melnyk’s conclusion also is important and timely because it comes at a time when the economic profession’s broad consensus about the damaging effects of minimum wages on employment has been challenged as part of the debate that preceded and influenced the recent decision to raise the minimum wage in the United States.

Since I found Melnyk’s technical analysis to be thorough and since I agree with his conclusions I will limit myself to some comments on the policy dimensions of minimum wage laws.

One way to start is to ask why anyone would want to impose minimum wages. In Poland, Melnyk tells us, the minimum wage had four main functions: (i) it was used by state enterprises to determine pay scales; and it was used by the government to (ii) calculate social benefits; (iii) to influence the overall level of wages; and (iv) to prevent poverty.

The first function, to assist state enterprises in setting the pay scales, should be rejected as being contradictory with the aim of reform–which is to allow relative prices and relative wages to be determined by market conditions and not by administrative formulae. The second function is unnecessary: social benefits can be targeted to specified personal income thresholds without any need to legislate a floor on wages. Indeed, Melnyk argues convincingly that the interaction between the minimum wage and the unemployment insurance system in Poland may well have induced individuals to voluntarily choose unemployment over work. The third function is to use minimum wages as an instrument of wage control, and this is an area where great damage can be inflicted.

When the process of transition to a market economy begins in earnest— in other words when prices are decontrolled— enterprises will find themselves in one of two categories: (i) those that produce goods for which demand has dropped (dramatically in those cases where prices had been kept well below equilibriumbrium levels before decontrol); and (ii) those that produce goods for which demand has increased. In a situation in which inter-enterprise movement of labor is costly (for psycho-sociological reasons, because of imperfections in housing markets or for other reasons), rigid sectoral wages will prevent the “losing” enterprises from reducing labor costs and thus will force an unduly large fall in output. Indeed, empirical evidence recently provided by Thierry Pujol ² strongly suggests that in Poland relative wage rigidity in the face of large changes in relative prices contributed significantly to the fall in employment and output. This is, of course, fully consistent with Melnyk’s results. The fact that minimum wages in Poland were so high, applied to such a large proportion of the labor force and were revised so often suggests that these legal minima contributed significantly to overall wage rigidity.

It is probably not a coincidence that in countries like Russia where minimum wages were revised less frequently than in Poland and were binding for a much smaller proportion of the workforce, the fall in output was accompanied with a much more limited rise in the unemployment rate. One possible explanation may be that in Russia the largely Party-affiliated labor unions were co-victims of the collapse of communist power and, with the exception of the coal miners, played virtually no role in economic decision-making in the subsequent reform period. In contrast, Polish unions played a key role in the demise of the communist system and thus an important role in the reform process. Indeed, Melnyk tells us that the minimum wage in Poland is no longer set by the government only but is determined jointly by the Ministry of Labor and the trade unions. This, of course, is like asking the bear to guard the barrel of honey.

The process of transition unavoidably involves a temporary drop in output. This is because within the group of enterprises for which demand falls there is a sub-group that produces goods that nobody wants and under the new set of relative prices these enterprises will have negative value added. The question faced by policy-makers is not how to avoid the contraction of the junk-goods producing sector, but how to facilitate a rapid redeployment of resources from this sector to more productive enterprises, thus making the fall in aggregate output as short and shallow as possible. This will require cutting subsidies and administrative credits to the junkproducing enterprises, but also eliminating barriers to inter-enterprise and inter-regional labor mobility, and removing those factors that reduce the incentives faced by profitable enterprises to expand the supply of jobs. Here again, the Polish minimum wage law appears to be an example of what not to do.

Let me now turn to the fourth alleged reason for minimum wages, which is to help the poor. Clearly, the minimum wage is a suboptimal way to achieve that goal. There is a way to redistribute income towards the poor without introducing a distortion in the labor market and without harming employment, i.e., through the tax system. A good example of that is the earned income credit in the United States, which is really a form of negative income tax. No one in the United States denies that this is a better system. Why then, have so many voted to increase the minimum wage? One answer that has been given is that a rise in the minimum wage does not increase unemployment. That’s the wrong answer. There is now a substantial body of empirical evidence — of which Melnyk’s paper is an example— indicating that minimum wages have significantly negative effects on employment whenever and wherever they are imposed.

The second answer is that the effect is small. Yet, a recent study by the IMF’s North American Division concludes that an increase in the U.S. minimum wage of 90 cents per hour (from $4.25 to $5.15 per hour)— as has been recently approved— would result in job losses of 100,000 to 300,000 workers. ³ This may be considered “small,” but if you are one of the job losers that will not do much to comfort you.

So, why minimum wages? Politicians sometimes have reasons that are difficult for reason to grasp. But I suspect that one reason is income redistribution— not from the rich to the poor, but rather from the outsider to the insider, from the unemployed to the employed, from the teenage non-voter to the adult voter.

Cuba’s future transition, like any process of economic transformation, will get special interest groups lobbying for privileges of all kinds, and it will be the task of economists to keep the global interests of society in mind. At some point pressures for minimum wage legislation probably will arise under the banner of social justice. Economists will then need to make it clear to politicians that they will be responsible for the increase in joblessness that will result, particularly among the youngest and less skilled members of society, and for its’ social and budgetary consequences. If the idea of a minimum wage cannot be killed, economists will have to argue for a system that minimizes adverse effects on employment— in other words a system that exempts the young (or at least assigns to them a lower wage floor, like in Canada); that is based on hourly earnings rather than on monthly earnings, unlike in Poland; and that keeps the gap between legal minimum and market-determined wages as narrow as possible.