CER | Unemployment: The case of Macedonia

Unemployment is by far the most damaging social scourge of the post-Communist economies in transition. Nothing seems to work. Even cycles of economic expansion seem to have the least effect on the labor markets. Macedonia is one of the worst hit.

The official figures are staggering: 37 percent of the workforce—about 300,000 people—are unemployed and looking for a job. Every 1.23 employees in Macedonia supports one unemployed person. In the USA the figure is between 3.3 and 4 employees supporting all the unemployed and all the pensioners!

The truth is less ominous. Many employed people in Macedonia go unreported as being such. Their employers prefer not to report them as employed in order to avoid paying social benefits and retirement benefits to the state. This greatly distorts the official figures—yet, it would be safe to assume that the unemployment rate in Macedonia is close to 25 percent. It is still inordinately high, socially corrosive and insupportable in the long term.

A certain level of unemployment is considered to be healthy. People move between workplaces—this is called labor mobility. People desert old professions for new ones, training themselves to occupy higher-paid positions that draw on higher levels of education. This kind of healthy unemployment is called "friction unemployment." A level of between three and six percent is considered to be friction unemployment in the West (depending in which country).

But the kind of unemployment that is prevalent in Macedonia (and in most other economies in transition) is not of this kind. It is permanent (structural), in the sense that the same people are unemployed continuously for more than a year. It is habit-forming: people lose their self-dignity, they become dependent on outside assistance, they are afraid to face reality. Such unemployment has grave psychological consequences. People change under its influence to such an extent that they no longer qualify as workers. Worse yet, people who used to provide for their families are cast aside as no-hopers, losers with no prospects for the future. This deeply and adversely impinges on the very fabric of society's basic unit: the family.

But unemployment also has a great macroeconomic impact. Macedonia doles out millions of Deutschmarks each month to pay unemployment benefits. Multiply DEM 60 to 100 per month by 300,000 job seekers and you see the frightening figures the Macedonian Minister of Finance is faced with every morning.

Instead of putting this money to productive use, it is spent on keeping people idle at home on an allowance which is not even enough for bare subsistence. Nobody is happy: the government, because its budget is unduly and unnecessarily stretched; the nation, because good money is thus spent instead of being invested and because scarce human capital is being squandered; and the unemployed, because they can hardly survive on what the state gives them.

Unemployment is not unique to economies in transition. Even much stronger economies, such as those of France and Spain, suffer from it. Spain's and the Netherlands's real unemployment rates used to be similar to Macedonia's.

A plethora of theories

What are the long term, structural causes for unemployment?

There are more theories than there are unemployed.

Some say that free trade encourages unemployment of unskilled and semi-skilled labor. Factories move overseas, where labor is cheaper. Inexpensive imports of textiles and basic electronic wares compete with the local production and, usually, wound it badly.

Others blame labor market rigidities. The laws and regulations of the state may favor a static workforce if, for instance, social benefits (annual vacation, sick pay, child support) increase the costs of employing workers, thus creating unemployment.

Furthermore, in inflexible job markets the psychology of employees and employers alike is that of "one big family" where no one is fired even in hard times and even if he or she is incompetent. Employers do not hire additional staff in times of economic boom, because they will not be able to fire them in time of crisis. They prefer to manufacture in places where labor costs are negotiable and low. Where trade unions have been abolished or significantly weakened (Britain and the USA are the prime examples), unemployment all but disappeared.

Still others emphasize the technological revolution (mainly in the fields of information technology). So many professions become obsolete at such a quick pace—and so many professions are revolutionized so often—that more jobs are lost than created.

The facts

We are all under the spell of magic words such as "mobility," "globalization" and "flextime." It seems as though we move around more frequently, that we change jobs more often and that our jobs are less secure. The facts, though, are different.

The world is less globalized today than it was at the beginning of the century. Job tenure has not declined (in the first eight years of every job) and labor mobility did not increase despite foreign competition, technological change and labor market deregulation. The latter led to an enhanced flexibility of firms and of hiring and firing practices (temporary or part time workers), but this is because many workers actually prefer casual work with temporary contracts to a permanent position.

Granted, people have been and are moving from failing firms and declining industries to successful ones and booming sectors. But they are still reluctant to change residence, let alone emigrate. Thus, jobs remain equally stable in deregulated as in regulated labor markets.

Yet, this phobia of losing one's job (arising from the aforementioned erroneous beliefs) serves to increase both the efficiency and productivity of workers and to moderate their wage claims.

It is safe to assume that collective bargaining led to increased wages and, thus, to less hiring and less flexible labor markets. It is therefore surprising to note that, despite the declining share of unionized labor in two thirds of the OECD countries, unemployment remained stubbornly high.

But a closer look reveals why. Both France and the Netherlands (where unionized labor declined from 35 percent of those actually employed to 26 percent), for instance, extended the coverage of collective agreements to non-unionized labor.

It is only where both union membership and coverage by collective agreements were both reduced (USA, UK, New Zealand, Australia) that employment reacted favorably. Thus, at the one extreme we find the USA and Canada where agreements are signed at the firm or even individual plant level. At the other pole, we have Scandinavia where a single national agreement prevails. All the rest are hybrid cases. Britain, New Zealand and Sweden decentralized their collective bargaining processes, while Norway and Portugal centralized it.

Damaging hybrids

The evidence produced by hybrid cases is not conclusive. Decentralized bargaining clearly reduced wage pressures, but centralized bargaining also moderated wage demands (union leaders tended to consider the welfare of the whole workforce). Still, it seems that it is much preferable to choose one extreme or the other rather than opt for hybrid bargaining.

The worst results, for instance, were obtained with national bargaining for specific industries. Hybrid Europe saw its unemployment soar from 3 to 11 percent in the last 25 years (and back to under 10 percent as it liberalized its labor markets). The pure market system in the USA maintained its low rate of four to five percent during the same quarter century.

These opposing moves cannot be attributed to monetary or fiscal policies. This is because all economic policies are geared towards increasing employment. Budget cuts, for instance, depress demand and job formation in the short term but, by lowering real interest rates, they encourage investment and job formation in the longer term.

The cycle is:

Employment protection laws make it hard to fire workers and hard for fired workers to find new jobs. The longer one is unemployed, the less the chances of finding employment. Skills rust and the long-term unemployed become the unemployable. Gradually, desperation sets in and the unemployed stop looking for a job. Their absence is conspicuous in that they do not restrain the wages paid to the employed. They have become part of the structural unemployment.

The weight of evidence

Blanchard and Wolfers studied 20 countries between the years 1960 and 1996. They applied eight market rigidities to their subjects. The average unemployment increased by 7.2 percent in this period. But in countries with strict employment protection, unemployment rose by double the amount experienced in countries with lax labor legislation.

The country with the most generous unemployment benefits saw its unemployment rate grow by five times the rate of the stingiest country. And in countries with highly coordinated wage bargaining, unemployment has grown by four times the growth in countries with decentralized bargaining.

It is difficult to isolate these parameters from the general decline in productivity, the increase in real interest rates and technological change and restructuring. Still, the results are fairly unequivocal.

Other research (the 1994 OECD one-year study, the DiTella-MacCullouch study) seems to support the thesis that flexibility is a good thing. It encourages employment, it leads to higher output and to a higher GDP per capita. The reason a transition from a rigid to a flexible labor market does not yield immediate results is that it increases the participation in the labor force. The rate of unemployment is, thus, affected only later, it lags the changes. But flexibility leads to lower rates of unfilled vacancies and to a lower persistence of unemployment over time.

Unemployment in Europe is structural (in Germany it has been estimated to be as high as 8.9 percent). It is the cumulative result of decades of centralized wage bargaining, strict job protection laws and over-generous employment benefits. The IMF puts structural unemployment in Europe at nine percent. This is while the USA's structural rate is five to six percent and the UK reduced its own from nine to six percent. The remedies, though well known, are politically not palatable: flexible wages, highly mobile labor and flexible fiscal policy.

Deregulation makes labor markets more flexible because it forces the worker to accept almost any job. Cutting or limiting jobless benefits has largely the same effect. Employers feel more prone to hire people if they can negotiate their wages with them directly and on a case-by-case basis and if they can fire them at will. Hence the debilitating effect of minimum wages and other wage controls as well as of job protection laws.

But all these steps must be implemented together because of their synergy. Research has demonstrated the impotence and inefficacy of half-hearted half measures.

Some hesitant steps have been adopted by the governments of Germany and France (which trimmed jobless benefits), by Italy (which stopped linking benefits to inflation), by Belgium, Spain and France, which reduced the minimum wage payable to young people. Spain established two classes of workers with an increased bargaining power granted to those with permanent employment. Yet, some measures yielded quite unexpected and unwanted results. France legislated a reduced working week. Other countries imposed a freeze on hiring with the aim of attrition of the workforce through retirement. Yet, these last two remedies led to an increase in the bargaining power of the remaining workers and to real wage increases.

The only clear causal relationship is between unemployment benefits and the level of employment. The lower the unemployment benefits, the more people seek work and wages decrease. As a result, firms hire more workers. But, firms hire even more when dismissing workers is made easier and cheaper.

Paradoxically, the easier it is to fire workers, the more workers firms are willing to take on and the more secure workers feel knowing that their chances of being hired are better. They look harder for work and find it, reducing the level of unemployment and the costs to the state of jobless benefits.

Having to spend less on unemployment benefits, the government can either cut taxes or improve the allocation of its resources. In both cases, the economy improves and provides an added incentive to work. This is because, in a vigorous growth economy, the value of an extra worker is higher than the combined costs of his hiring and firing.

This is especially true since the reservoir of the unemployed is comprised of the unskilled, the young and women, whose remuneration is closer to the minimum wage. In the USA, the minimum wage is 35 percent of the average wage (in France it is 60 percent, in Britain it is 45 percent and in the Netherlands it is declining relative to the median salary). It is a fact that when wages are downward more low-skilled jobs are created. A one percent rise in the minimum wage reduces the probability of finding a job by between 2 and 2.5 percent.

A poverty paradox

There is a debate raging between the proponents of minimum wages (they reduce poverty and increase the equality of wealth distribution) and their opponents (they destroy jobs). The OECD stated clearly that wage regulation could not deal with poverty. The reason is that, as opposed to common opinion, few low-paid workers live in low-income households and few low-income households have low-paid workers. Thus, the benefits of the minimum wage, such as they are, largely bypass the poor.

Again, it is important to realize that unemployment is not a universal phenomenon. It is concentrated among the young and the unskilled. 11 percent of all people under the age of 25 in the USA are unemployed, almost three times the national average. A shocking 28 percent of those under the age of 25 are unemployed in France. The OECD says that a ten percent rise in the minimum wage reduces teenage employment by two to four percent in both the high and low minimum wage countries.

In view of these facts, many countries (USA, UK, France) introduce "training wages"—actually, minimum wage exemptions for the young. But the minimum wage is still a high percentage of mean youth earnings (53 percent in the USA and 72 percent in France) and thus has a prohibitive effect on youth employment.

There is no disputing the facts that minimum wages compress the earnings distribution and reduce wage disparities between ages and sexes, but they have no effect on inequality and the reduction of poverty among households. In US households with less than half the median household income, only 33 percent of adults have a low-paid job (the equivalent figure in the Netherlands is 13 percent and in the UK it is five percent).

In most poor households, no one is employed at all. On the other hand, many low earners have high-paid partners. In the USA only a third of earners of less than two thirds of the median wage live in households whose income is less than half of the national median household income. In the UK the figure is ten percent and in Ireland three percent. In each five-year period only a quarter of low-paid Americans are in a poor family at some point (the figure is ten percent in the UK).

These statistics show that minimum wages hurt poor families with teenagers (by making teenage employment prohibitive) while benefiting mainly the middle class.

Unemployment and inflation

Another common misconception is that there is some trade-off between unemployment and inflation. Both Friedman and Phelps attacked this notion. Unemployment seems to have a "natural" (equilibrium or homeostatic) rate, which is determined by the structure of the labor market. The natural rate of unemployment is consistent with stable inflation (NAIRU – Non Accelerating Inflation Rate of Unemployment).

Making more people employable at the prevailing level of wages can lower NAIRU. This should lead to a big drop in unemployment, together with a tiny increase in permanent inflation. Phelps actually sought to lower NAIRU and raise the incomes of the working poor. Stiglitz calculated that the changing demographics of the labor force and the competition in markets for goods and jobs reduced NAIRU by 1.5 percent in the USA. R Gordon, D Staiger and M Watson support these findings.

It emerges, therefore, that the gap between the estimated NAIRU and the actual rate of unemployment is a good predictor of inflation.

Going Dutch

The Anglo-Saxon variety of capitalism is intended to maximize value for shareholders (often at the expense of all others, including the workers).

The Rhineland model is capitalism with a human face. It calls for an economy of consultation among stakeholders (shareholders, management, workers, government, banks, other creditors, suppliers, etc). In the Netherlands, there is a Social and Economic Council. Its role is advisory and it is semi-corporatist. Another institution, the labor Foundation is a social partnership between employees and employers.

But the Netherlands succeeded in reducing its unemployment rate from 17 percent to less than 5 percent by ignoring both models and inventing the "Poldermodel," a third way. Wim Duisenberg, the Dutch banker (currently Governor of the European Central Bank), attributed this success to four elements:

Improving state finances
Pruning social security and other benefits and transfers
Flexible labor markets
Stable exchange rate

The Dutch miracle started in 1982 with the Wassenaar Agreement, in which employers' organizations and trade unions agreed on wage moderation and job creation, mainly through decentralization of wage bargaining. The government contributed tax cuts (which served to replace forgone wage increases). This fiscal stimulus prevented a drop in demand as a result of wage moderation.

Additionally, restrictions were placed on social security payments and the minimum wage. For instance, increases in wages were no longer matched by corresponding increases in minimum social benefits. Working hours, hiring, firing and collective bargaining were all opened up to labor market forces. The strict regulation of small and medium size businesses (which drove up labor costs) was relaxed. Generous social security and unemployment benefits (a disincentive to find work) were scaled back.

The Netherlands did not shy from initiating public works projects, though on a much smaller scale than France, for instance. The latter financed these projects by raising taxes and by increasing its budget deficit. The result of this course of action, though, could well be a reduction in employment in the long run (the effect of taxation). In the absence of monetary instruments, such as devaluation (due to the EMU), the only remedy seems to be labor market flexibility.

Such flexibility must include a substantial adjustment in sickness benefits, vacation periods, maternity leave and unemployment benefits.

Introducing flexibility

The long-term (more than 12 months) unemployment in Europe constitutes 40 percent of the total unemployment. About half of the entire workforce under the age of 24 is unemployed in Spain. It is about 28 percent in France and in Italy. Germany, Austria and Denmark escaped this fate only by instituting compulsory apprenticeship. But the young become the kernel of long-term unemployment. This is because a tug of war, a basic conflict of interests exists between the "haves" and "have-nots." The employed wish to defend their monopoly and they form labor cartels. This is especially true in dirigiste Europe.

While in the USA 85 percent of all service jobs created between 1990 and 1995 paid more than the average salary, this was not the case in Europe. Add to this the immobility of labor in Europe and a stable geographical distribution of unemployment emerges, not ameliorated by labor mobility.

The Dutch model sought to battle all these rigidities:

The Dutch reduced social security contributions from 20 percent (1989) to 7.9 percent and they halved the income tax rate to 7 percent (1994).
They allowed part time workers to be paid less than full timers, doing the same job.
They abandoned sectoral central bargaining in favor of national bargaining—but more decentralized.
They cut sickness benefits, unemployment insurance (benefits) and disability insurance payments (by ten percent in 1991 alone).
They made it harder to qualify for unemployment benefit (in 1995 no benefits were paid to those who chose to remain unemployed).
The burden of supporting the sick was shifted to the employer / firm. In 1996, the employer was responsible to pay the first year of sickness benefits.

Even the Dutch model is not a success. More than 13 percent of the population is receiving disability benefits. Only 62 percent of the economically active population is in the workforce (the rest dropped out of it).

The failure of reform

But compare its experience to France, for instance.

The Loi Robien prescribes that companies should be spared social security obligations for seven years if they agree to put workers on part-time work instead of laying them off. Firms abused the law and restructured themselves at the government's expense.

The next initiative was to reduce the working week to 35 hours. This was based on the "Lump of Labor Fallacy," the idea that there is a fixed quantity of work and that reducing the working week from 39 to 35 hours will create more jobs. In reality, though, labor demand changes only in response to changes in productivity and in the workings of the labor market itself (rigidities). A cut in the working week reduces productivity and destroys jobs rather than fostering job formation.

In Spain, a permanent employee fired is entitled to receive up to 45 days' pay multiplied by his or her tenure in years. The result is that firms are afraid to hire or fire workers. The government, faced with more than 22 percent unemployment, permitted part-time contracts with less job protection. Today, 30 percent of all employed Spaniards work this way. Yet, this led to the creation of a two-tiered workplace where it is easier to fire the part-timer (even if he is valuable) rather than the permanent (and better earning) worker. Additionally, wages are thus disconnected from productivity.

Absorbing migrants

But whatever the reasons are for unemployment, certain countries are battling this cancer of society in creative ways.

During the 1990s, Israel—a country with 4.5 million people and 20,700 square kilometers—absorbed an inflow of more than 600,000 immigrants (ie 15 percent of the population), mainly from the former USSR.

One would expect a dramatic increase in unemployment. If Macedonia were to absorb 15 percent of its population—300,000 additional immigrants—tomorrow, its unemployment rate would have skyrocketed until the newcomers had been absorbed by the marketplace. This was made painfully evident with the influx of Kosovar refugees during the 1999 Operation Allied Force.

But not only did Israel succeed in providing most of this deluge of immigrants with jobs, it also reduced the overall rate of unemployment among its old population! How did it succeed in doing the impossible? It is an example closely studied by most economies in transition.

Sam Vaknin, 29 January 2001

Part two of this series of six articles will appear in next week's CER

The author:

The author is General Manager of Capital Markets Institute Ltd, a consultancy firm with operations in Macedonia and Russia. He is an Economic Advisor to the Government of Macedonia.

DISCLAIMER: The views presented in this article represent only the personal opinions and judgments of the author.

Moving on:

After the Rain cover

After the Rain:
How the West Lost the East

Sam Vaknin's book on sale from CER as a print book and as an ebook
Read an external review