Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 177
Transforming Post-Communist Political Economies 7 Formal Employment and Survival Strategies After Communism Simon Johnson, Daniel Kaufmann, and Oleg Ustenko INTRODUCTION The countries of Eastern Europe and the former Soviet Union have suffered large declines in output and, in most cases, employment since the collapse of their communist regimes. Most research on labor markets in these economies has focused on changes in formal employment status. Numerous studies, using techniques developed for Western market economies, have examined the flows into and out of formal jobs, particularly in Eastern Europe (see, for example, all the papers in Commander and Coricelli, 1995; Coricelli and Revenga, 1992). The assumptions underlying this work are clearly presented in the leading labor market model for post-communist countries (see Aghion and Blanchard, 1994; Blanchard et al., 1995). According to this model, the beginning of economic reform means a jump downward in state-sector employment, followed by a steady decline. There is a net creation of jobs in the private sector, but not enough to absorb all the people who leave the state sector. The result is steadily increasing unemployment. People who remain employed or who find new jobs do relatively well, while those who are unemployed, particularly for a long time, do badly. In this conventional model, state enterprises either keep people employed or fire them. Individuals respond to being fired by either finding a new job or not. Some individuals may also quit and move directly to the private sector. The model is therefore a sequence of discrete, zero-one choices by both firms and individuals. This work was partly supported by the World Bank. The views are the authors' and do not necessarily reflect those of the institution.
OCR for page 178
Transforming Post-Communist Political Economies For post-communist economies, this model is not appealing for three reasons. First, there is a great deal of anecdotal and survey evidence indicating that firms can vary greatly the intensity with which they use labor. Standing (1994) documents what he calls "hidden unemployment" in several forms. Hours worked can be reduced through closing the factory for some period, through sending people on forced vacation (paid or unpaid), and through offering extended maternity leave. A person can also be paid a lower wage, either relative to others or in absolute terms, with it being understood on both sides that this implies a lower level of time commitment to the firm. Second, individuals can also respond to changes in their formal employment by engaging in various "survival strategies," or alternative ways of earning income, primarily through informal self-employment. In principle, an individual can vary his or her intensity of work in these survival activities over an almost continuous scale. Third, it seems inappropriate to assume that firms make one formal employment decision per person, and each individual then decides how to respond. Rather, the interaction between most individuals and their original employer is repeated, with both sides taking into account both the previous and expected future actions of the other. For example, if the managers of a firm see an individual heavily engaged in survival strategies, they may consider cutting back on that individual's work in the firm, although not necessarily firing him or her. The individual may respond by putting more effort into other income-earning opportunities, but not necessarily quitting. Our empirical work is guided by a theoretical framework that modifies the standard model to take these three considerations into account. We model the individual's decision as a stochastic dynamic supermodular optimization problem. This problem has two critical assumptions. First, we require that managers choose the intensity with which each individual works in the firm, while the individual chooses the intensity with which he or she works outside the firm. As the manager reduces the intensity of firm work, there is a higher payoff to the individual from increasing the intensity of outside work. The converse may also be true: if the individual works more outside the firm, the manager derives a higher payoff from reducing the intensity of work for this person. The second critical assumption is that larger adjustments are more costly than smaller ones for both firms and individuals. This slows down the optimal speed of adjustment on both sides and means that optimal intensities of work adjust over time (although the optimal speed can vary considerably across individuals). These two assumptions describe a supermodular optimization problem. Static versions of supermodular problems can be analyzed using the tools provided by Milgrom and Roberts (1990), but this approach does not capture important aspects of transitional economies. Instead, we use the results established by Friedman and Johnson (1996), which show that in the presence of complementarities and convex-type adjustment costs, the dynamic adjustment
OCR for page 179
Transforming Post-Communist Political Economies path for the individual increases in parameters that change the overall economic environment. The appeal of this approach, as shown by Friedman and Johnson (1997), is that it allows us to generate robust testable hypotheses.1 These results provide a framework that is consistent with a variety of specific models about interactions between firms and individuals, and that encompasses important elements of established theory about Eastern European labor markets as a special case. This is an attractive feature because while there is no agreement on precisely how firms restructure and how individuals behave in the post-communist economies, aspects of most existing models can be formulated in terms of our supermodular framework. Our evidence is drawn from a survey of 1828 current and former state enterprise workers, conducted during the summer of 1994 in Ukraine. The sample comprises people who were separated, people identified by management as working less than normal hours or earning unusually low wages, and people who were randomly selected from among the employed. The workers were previously or still employed at 26 organizations, drawn from a cross-section of sectors and based primarily in the heavy industrial region of eastern Ukraine, but including some research institutes and schools in addition to firms. With this evidence, we can assess the standard model of labor market adjustment. We find evidence of much more general complementarities than are predicted by the conventional theory. Using both direct measures and estimates of earnings, we find evidence (from cross-tabulations) that people who worked less inside the firm worked more outside the firm, and vice versa. However, in simultaneous regression estimation, much stronger results are found for demographic variables. In effect, we find that while most people can offset poor formal employment outcomes through use of survival strategies, women and pensioners are less able to do so. These groups appear significantly less able to cope by earning income in the private sector. Another impressive result is the extent to which people across the board earn at least a survival level of income through informal or secondary activities. In our sample, 70 percent of the randomly selected employed people were engaged in a survival strategy of some kind. Remarkably, we find that people who remained employed were more likely to engage in survival strategies than were those who were unemployed; the proportions are 69 percent for the employed and 65 percent for all the unemployed (56 percent of the unemployed receiving assistance and 79 percent of the unemployed not receiving assistance).2 Using the lowest plau- 1 Friedman and Johnson (1997) show that for a wide range of dynamic optimization problems, supermodularity is both necessary and sufficient for monotone static results. In the present context, this implies that our supermodular model requires the minimum set of assumptions to obtain monotonicity in the optimal decision variables. 2 The evidence presented here needs to be supplemented with information about inter- and intrafamily income transfers. This issue was addressed in a follow-up survey, but analysis of the results is not yet complete.
OCR for page 180
Transforming Post-Communist Political Economies sible estimates of income from survival strategies, employed people earned an average of $15.9 per month in their primary formal employment and $29.3 per month through other activities. At that time, the cost of a minimum survival basket of food per adult was at least $20 per person.3 Many people therefore appear to have responded to worse outcomes in their formal employment with various forms of survival strategies. These results extend three literatures. The first is recent work by Western labor economists on a range of post-communist countries (Commander and Coricelli, 1995). Using an approach developed in the study of Western labor markets, this literature has focused on formal employment and the measurement of ''true" unemployment. A major puzzle for this theory is that even though the employment decline has been of a similar order of magnitude, unemployment rates in the former Soviet Union are much lower than those typical in other areas of Eastern Europe. Registered unemployment in Eastern Europe is over 10 percent everywhere except in the Czech Republic, and up to 20 percent in some countries. In the countries of the former Soviet Union, it is generally less than 5 percent, and in Ukraine during 1994 it remained below 1 percent. Layard (1994) argues that Russia has had more labor market flexibility, including lower real wages, but he does not model secondary employment strategies. While supporting the finding that there are problems for some of the unemployed in post-communist labor markets, our results are perhaps closer to the literature on the second economy during and particularly at the end of the Soviet period (see, for example, Alexeev and Treml, 1993). According to this work, the second economy was always strong, and economic reform brought to the surface activities that previously existed but had been hidden (Leitzel, 1995). Consistent with this argument, we show how people continue to use second-economy activities to survive after the collapse of communism. Third, we also show how to estimate empirically a general model of strategic complements in labor market adjustment. The theoretical work of Milgrom and Roberts (1990, 1994) has not yet been used extensively in econometric work. Our results indicate that with the modifications provided by Friedman and Johnson (1996), this approach offers an extremely attractive general framework for testing rival hypotheses. The next section summarizes the Ukrainian economic situation through the summer of 1994. We then develop our model of strategic complements between firm and individual decisions concerning formal employment and survival strategies. The following two sections, respectively, present tests based on directly measuring the intensity of work and use income as a proxy for intensity. Next we examine what determines whether a person is coping or not. The final section presents conclusions and policy implications. Annex 7-1 contains a detailed assessment of gross separations in our sample. 3 The cost of this food basket is based on our own food price surveys, which are now used in Ukranian Ministry of Economics publications (e.g., Zienchuk, 1994-1995).
OCR for page 181
Transforming Post-Communist Political Economies THE UKRAINIAN ECONOMIC SITUATION4 By the summer of 1994, Ukraine was in a deep crisis. With the breakup of the Soviet Union, the country suffered a major deterioration in its terms of trade, particularly with Russia and Turkmenistan, which provide most of Ukraine's energy imports. Its share of gas and oil imports by value increased from 12 to over 50 percent of imports from the former Soviet Union between 1991 and 1993. This increase in the imported energy bill was met by compressing nonenergy imports—by about 35 percent a year in 1993 and 1994—and by increasingly running energy arrears to Russia and Turkmenistan. As the result of an overvaluation of the relevant exchange rate, exports fell by over 15 percent a year in both 1993 and 1994. From 1990 to 1994, official gross domestic product (GDP) fell by over 50 percent, and measured in dollar terms at the market exchange rate reached around $33 billion ($600 per capita). In purchasing-power terms, Ukrainian GDP was approximately $3500 (in 1994 dollars). Inappropriate economic policies contributed significantly to this decline in economic activity. Throughout the postindependence period, the state maintained, and at times further tightened, administrative controls over economic activity, particularly the exchange rate, domestic and external trade, and prices. In addition, there were frequent changes in regulations. This regulatory environment hindered restructuring and the development of the private sector and induced many activities to become unofficial or go underground, thus eroding the tax base (Kaufmann, 1994). Loose fiscal and monetary policies were largely responsible for the very high inflation Ukraine experienced after 1991 (Åslund et al., 1994). Monetary financing of the budget deficit was between 10 and 25 percent of GDP, while the average inflation rate climbed from 90 percent in 1991 to 1,210 percent in 1992 and to 5,000 percent in 1993. By the fall of 1993, inflation rates were approaching hyperinflation, with prices rising about 65 percent a month. Following a tightening of credit policies and a reinstatement of administrative controls (such as on energy and other prices), the rate of inflation fell in 1994, although it was still estimated to be about 800 percent for the year. The sharp decline in production and rapid increase in prices significantly eroded the standard of living of the population. There was a marked deterioration in schools, hospitals, and health care more generally. By the summer of 1994, almost all state enterprises had encountered serious financial problems and faced the need to cut employment or wages or both. Realizing the extent of the crisis, the new president, Leonid Kuchma, who was elected in mid-1994, called for a break from previous economic policies. 4 This section draws on various World Bank documents, particularly the Report on the Rehabilitation Loan, November 1994, as well as Kaufmann (1994), Johnson and Ustenko (1995), and Åslund et al. (1994).
OCR for page 182
Transforming Post-Communist Political Economies A comprehensive program of macroeconomic stabilization and structural reforms was developed, and initial measures to stabilize, liberalize, and privatize the economy were initiated in late 1994. The results presented below have not been influenced by these new policy developments; instead our evidence reveals formal employment and survival strategies in the crisis months of the summer of 1994. THE MODEL The two critical assumptions presented in the introduction concern the complementarities between the actions of firms and individuals and the nature of adjustment costs. Two actions are complementary if taking one does not preclude taking the other, and taking both together gives at least as much benefit as taking the two separately.5 In the case of differentiable functions, complementary is identical to positive cross-partial derivatives. Intuitively, this means doing more of one activity increases the benefit from doing more of another activity.6 Here we model the potential complementarity between the intensity with which managers employ individual workers inside the firm and the intensity with which workers engage in income-earning activities outside the firm. The firm's employment strategy, fti, is controlled by management and represents the intensity with which the firm employs this worker. If fti is higher, individual i is working less intensely inside the firm; that is, the degree of "underutilization" of this person has increased.7 A higher value of fti means fewer hours worked, less salary paid, and in general less intense work for a particular person. Nothing in our model requires that fti or any other variables be continuous or differentiable. In fact, we would expect many of these variables—such as whether the worker is given a "forced" vacation—to be discrete in nature. The individual chooses it, the intensity with which he or she works outside the firm. There is also an adjustment cost, which depends on the magnitude of the change in his or her action, it, between periods. This variable is 5 Formally, we require players' strategy sets to be compact sublattices. This represents the idea that individuals' activities can be ranked in an ascending order and that any combination of activity levels is feasible. Each activity is totally ordered, but when they are considered together, the resulting vector is only partially ordered. The lattice assumption does not require continuity or convexity in the activities, so discrete and continuous variables can be treated in the same way. Precise formal requirements are explained in Friedman and Johnson (1996). 6 Our theory, and the work on supermodular functions more generally, assumes only weak complements; that is, the conditions are all satisfied if there is no cross-effect. The only requirement is that the cross-partial derivatives (or the appropriate analogy for nondifferentiable functions) be non-negative. 7 We assume there is a single decision maker in each firm who controls all the elements off,, so we can use "firm," "manager," and "management" synonymously.
OCR for page 183
Transforming Post-Communist Political Economies intended to capture the idea that it takes time for individuals to develop outside opportunities, so more rapid adjustment is more costly. Given these versions of our two critical assumptions for each individual's payoff function (b) and adjustment cost function (I), the overall objective function is also supermodular. (For formal proofs of these points, see Friedman and Johnson, 1996.) The individual's objective function can be written as the present discounted value of payoffs minus adjustment costs: (1) where fti is the firm's action and it is the individual's action, and w is a vector of parameters describing the individual from the individual's point of view (and taken as given by all decision makers). The single period payoff function is given by while adjustment costs are I(it , it+1). We will test whether a lower intensity of work inside the firm is complementary to a higher intensity of work outside the firm, that is, whether working less inside the firm raises the payoff to working more outside the firm. In this framework, a higher intensity of reform policy leads to a greater reduction in the intensity of work inside the firm and an increase in the intensity of work outside the firm (but not necessarily for all individuals). We can now formulate the two hypotheses to be tested. The first is as follows: Hypothesis 1. Controlling for individual characteristics, the intensities of inside and outside work are negatively correlated (i.e., a lower intensity of inside work implies a higher intensity of outside work, all other variables being held constant). If this hypothesis is correct, the intensity of work inside the firm should be negatively correlated with the intensity of work outside the firm. For example, the individuals who work more intensely outside the firm (using what we call "survival strategies") should be those who work less inside the firm. Some of them may leave the firm altogether, as implied by the conventional theory, but our model also allows other forms of reduction of work intensity short of complete separation. Note that nothing in our model requires that every person work at a reduced intensity inside the firm during or after the completion of reform. In fact, our empirical work in Russia and Ukraine indicates that some people continue to work at a high intensity while others are gradually pushed out of the firm. We also assume that firms perceive differences among individual workers. Worker attributes are represented in the vector w, where a higher value means that the firm has a higher incentive to reduce that individual's intensity of inside work. This may be because that worker is currently paid more than his or her marginal product, or because the worker is a troublemaker, or
OCR for page 184
Transforming Post-Communist Political Economies because the manager feels he does not need to employ this kind of worker because she has excellent outside opportunities. Again, an important advantage of our framework is that all these specific explanations are consistent with our formal model. Hypothesis 2. Individuals' characteristics significantly affect the intensity of inside and outside work, even when we control for potential simultaneity. Our empirical work therefore includes regression analysis with the intensity of inside and outside work on the left-hand side in a pair of simultaneous equations. Variables on the right-hand side will include demographic and firm-specific information. We follow this analysis with a third testable hypothesis implied by, but not formally derived from, our theory: Hypothesis 3. People who do not earn a minimum survival level of income (i.e., who do not cope on their own) are those who work significantly less both inside and outside the firm. THE INTENSITY OF WORK Testing our first two hypotheses requires an analysis of the intensity of work both inside and outside the firm. This section provides direct measures of intensity, while the next uses income earned as an indicator of intensity. The following section deals with the third hypothesis, about people who can and cannot cope. The Sample Our sample contained 26 enterprises, of which 19 were from Dnepropetrovsk, 3 from Kiev, and 4 from Simferopol (in Crimea). Our goal was to cover a wide cross-section of the economy, and to this end we included 2 schools, 4 industrial institutes (which provide design and research support for manufacturing and construction), 5 machine-building enterprises, 2 equipment-building enterprises, 3 food-processing enterprises, 3 chemical enterprises, 4 enterprises producing construction materials, and 4 enterprises producing consumer goods. We wished to contact 75 people from each organization, 50 of whom were still employed and 25 of whom had separated. In each organization, we obtained a list of workers from the personnel office. Because we were particularly interested in the behavior of low-income people, 25 of those still employed were to be people identified by the person-
OCR for page 185
Transforming Post-Communist Political Economies nel office as being "unsuccessful," which our interviewers defined as meaning working fewer than normal hours, having taken a forced vacation, or receiving unusually low wages. Our intention was to have management identify some of the people who could be considered in some sense to be "hidden unemployed." The other 25 employed people were randomly selected from among the entire staff of the organization, with the interviewer being told to visit as many departments as possible. Note that our sample probably underrepresents upper-level managers. We also wanted to examine differences between people who remained employed and those who had separated. We therefore obtained from management a list of people who had separated. We interviewed people on this list by telephone until we had responses from 25 people or until the list was exhausted.8 Our final sample comprises 1828 people, 30 percent of whom were randomly selected, 37 percent of whom were employed and on the list provided by management, 14 percent of whom were separated and employed, and 19 percent of whom were separated and unemployed. Our sample therefore represents the three groups we wanted to investigate in roughly equal proportions. Compared with the total population of people employed in these firms in 1991, of course, it overrepresents both those employed with fewer than normal hours and those who were separated.9 Sampling in this way enabled us to draw interesting comparisons across employment categories, but it means one must be careful about using the sample to infer overall proportions in the total population. For the most part, the same questions were asked of the separated and nonseparated. However, the separated were also asked some additional questions about what happened to them after separation (e.g., whether they received unemployment assistance). Use of Survival Strategies Our survey asked whether a person was engaged in any of six types of survival strategy: "another job" (this was asked only of people we knew had at least one formal-sector job, i.e., who had not separated or who had separated and were reemployed), use of a dacha or other plot of land to grow food, work to some extent as a private taxi driver, renting out of one's apartment, business trips abroad, and renting out of a garage. With the exception of the rentals, all 8 Cost considerations precluded us from interviewing separated people in person. However, we do not think the difference between phone and in-person interviews biases our results. 9 About 6 percent of those still employed earned the minimum wage. However, estimates of "hidden unemployment" range as high as 34 percent. The specific size and nature of the "unsuccessful" group depends on the precise definition used.
OCR for page 186
Transforming Post-Communist Political Economies the questions asked here related to informal employment (although answers about "another job" could refer to another formal job).10 For the dacha question there were four possible answers: "I don't have a dacha," and "I grow food for myself a little," ". . . for myself a lot," or ". . . to sell.'' We considered only the second and third of these answers to represent the "dacha survival strategy." Almost everyone we know in Ukraine who has a dacha grows some food for his or her family and always has. We therefore regard the response ". . . for myself, a little" as being a culturally determined income-supplementing activity that will probably always exist in Ukraine and should not be confused with survival strategies due to the difficult economic situation. We also asked whether a person worked in any way "as a taxi driver." There is an established tradition in Ukraine of giving people rides for short distances (effectively hitchhiking in the city). Prices for this service are high compared with official state-sector wages—about $1 equivalent for a 10-minute ride in Kiev at a time when state-sector wages are $20 per month. There were three possible answers: no, part-time, and full-time. Anyone who gave either of the latter two answers was classified as using the taxi survival strategy. We asked people further whether they rented out their apartment or shared in the income of another family member who rented his/her apartment (and as a result lived with the respondent). We were also interested in what foreign business trips people had taken in the last year. Specifically, we asked whether people had traveled to Russia, Poland, Turkey, China, or Western Europe.11 Answers were classified as one to three, four to seven, or more than seven trips to each destination. Finally, we asked people whether they rented out their garages. Of the fully employed, 5 percent did so. Table 7-1 shows a remarkably high use of survival strategies across the board. Of all the employed (nonseparated) people in our sample, 70 percent made some use of at least one strategy. People working fewer than 30 hours per week made as much use of the strategies as did the unemployed who did not receive assistance (which indicates some support for Hypothesis 1). However, even people working at least 40 hours a week were only slightly less likely to have a survival strategy. Of the 493 people in our sample who had a car (34 percent of the total), 34 percent worked as private taxi drivers (4 percent said they worked full-time in this occupation). This survival strategy shows the biggest difference between the sexes, being cited by 19 percent of men, but only 2 percent of women. 10 In this survey we did not ask about income transfers. The survival strategies here are therefore "active" ones, involving some work, but are only a subset of all strategies. Our follow-up survey deals with this issue in more depth. 11 We did not ask about trips to the Baltic states or other former Soviet Union countries. The numbers here therefore understate the amount of foreign travel.
OCR for page 187
Transforming Post-Communist Political Economies TABLE 7-1 Proportion of People Engaged in Various Survival Strategies (all numbers are percentages) Dacha Trips Taxi Rent Second Job ≥1 Strategy Unemployed with assistance 14 26 4 26 n.a. 56 Unemployed without assistance 40 40 8 26 n.a. 79 Of nonseparated wage earners: Less than 20 hours 37 21 19 15 22 74 20 to less than 30 26 36 10 15 33 79 30 to less than 40 33 28 21 9 26 68 40 or more hours 25 29 12 17 20 69 All sample 24 25 11 13 20 70a a This percentage refers to the employed only and not the whole sample. This is obviously related to the fact that 78 percent of women, in comparison with 56 percent of men, said they did not have a car. Of the 1828 respondents, a striking 503 (28 percent) had taken a trip of some kind. Most of these people (15 percent of the sample) had gone to Russia three or fewer times, while 5 percent had gone there more than three times; 81 (4 percent) had gone to Poland, 35 to Turkey, and 12 to China. The pattern is interesting. Older people, women, and the unemployed who were not receiving assistance were more likely to go to Russia and less likely to travel to non-Russian destinations. At this informal level of analysis, we find some support for our first hypothesis. Cross-tabulations show survival strategies are used somewhat more by people who work less in formal jobs. We now test more formally for the effect of work intensity on the use of survival strategies. Regression Results Table 7-2 shows regression results for the simultaneous equation system in which the inside and outside intensity of work are the left-hand-side variables. The inside intensity is measured as the number of hours per week the respondent reported working in the firm, while the outside intensity is an index measuring the extent to which an individual engaged in survival strategies. In this index, each basic use of a strategy is given 1 point, while each use above the minimum is given an additional point. The index is capped at 5 points.12 All regressions were estimated by three-stage least squares, using instrumental variables and taking into account the simultaneity. The instruments were 12 It made no difference to our results when the index was capped at 8 points.
OCR for page 192
Transforming Post-Communist Political Economies TABLE 7-4 Percentage of Secondary Incomes in Total Income Secondary Incomes in Total (percent) Index of Strategies (average) Outside Earnings (average) Unemployed without assistance 100 $27.0 Wage distribution 1st quintile 77 1.65 38.5 2nd quintile 60 1.52 23.7 3rd quintile 44 1.43 26.4 4th quintile 44 1.33 20.3 5th quintile 29 1.47 20.3 All sample 54 formal job tended to work more (as measured by our survival strategy index) and earn more through outside work. Again the cross-tabulations support Hypothesis 1. Also striking is the absolute magnitude of the compensation. In Table 7-3 we see that average total incomes were twice as high as official formal-sector wages for almost all groups, and three times as high for some groups. On average, people could more than compensate for low official incomes. Equally remarkable is the finding about the total earnings of unemployed people (last two columns of Table 7-3). Not surprisingly, unemployed people receiving assistance had low incomes, although even without considering benefits they averaged more than the randomly selected employed group earned officially. More striking, however, is the high average incomes of people who were separated and were not receiving assistance; this average is actually higher than for people who were separated and employed. Average incomes can, however, be misleading, particularly for the unemployed. Table 7-5 shows the distribution of total earnings by employment status. The lowest part of the income distribution is significantly lower for unemployed people without assistance than it is for the randomly selected nonseparated people, but the medians are much closer together. This is because people could earn large amounts from survival strategies relative to official earnings, but for some reason, 30-40 percent of the unemployed did not engage in these strategies. Thus there was a very large dispersion of incomes among those unemployed without assistance. Although on average survival strategies allowed people to compensate for low formal earnings, this was not true for everyone. The lowest quintile of wage earners earned less than $9.3 per month in formal employment. The lowest quintile of the income distribution (including all survival strategies) earned less than $14.3 per month, and the lowest one-third earned less than
OCR for page 193
Transforming Post-Communist Political Economies TABLE 7-5 Distribution of Total Earnings by Employment Status (indollars at market exchange rate) Randomly Selected Nonseparated Separated and Reemployed Unemployed with Assistance Unemployed Without Assistance Minimum 4.5 9.2 0 0 10th 415.2 14.8 0 0 20th 18.5 17.4 0 3.3 25th 19.7 18.2 0 5.0 30th 20.9 19.1 1.7 5.8 40th 23.9 22.1 3.3 10.7 60th 30.7 25.6 5.0 28.3 70th 40.2 29.9 10.0 38.3 75th 52.4 35.0 25.0 45.0 80th 60.6 38.5 25.0 45.8 90th 63.3 42.6 26.7 46.7 Median 74.3 66.3 45.0 115.0 Max 422.1 167.6 163.3 165.0 $17.2 per month. Of the lowest quintile of wage earners, 45 percent are in the lowest quintile, and 50 percent are in the lowest one-third of the income distribution. In conclusion, differences in average total income are much smaller than those for average formal wages (supporting Hypothesis 1). Average earnings from survival strategies are the highest for the unemployed without assistance, while the lowest such earnings are for the reemployed—the group with the highest average official wages. Given that we use the lowest possible estimates of survival strategy earnings, we are showing the smallest plausible equalization effect. The effect is large primarily because official earnings were so low. In this situation, it does not take much in the way of survival strategy earnings to have a large compensating effect. However, not everyone could compensate effectively. The income distribution within each formal employment group actually widens when informal incomes are included. To see which people were more or less able to compensate for low earnings in their regular jobs through informal activities, we need to use multivariate regression analysis. Regression Analysis Table 7-6 shows the results from simultaneous equation estimation. The specifications and estimation procedures are the same as reported in Table 7-2, but now the left-hand-side variables are our estimates of inside and outside earnings. The first two columns are only for the nonseparated workers, while
OCR for page 194
Transforming Post-Communist Political Economies TABLE 7-6 Regression Results for Inside and Outside Earnings 1. Only Nonseparated 2. Nonseparated and Unemployed Inside Earnings Outside Earnings Inside Earnings Outside Earnings Constant 16.0* 19.7 14.42* 8.3 (1.63) (13.7) (1.58) (12.1) 18-25 years old −3.90* −6.38 −82.99* −4.05 (0.93) (5.4) (1.02) (4.8) 26-35 years old −1.31 3.47 −0.51 4.49 (0.74) (4.2) (0.83) (3.76) 36-45 years old 0.50 −0.64 0.80 −0.47 (0.71) (3.9) (0.81) (3.73) Dummy form pension age −3.98* −8.72 −2.35 −7.49 (1.23) (6.78) (1.44) (6.22) Female dummy −2.59* −0.38 −2.28* −1.70 (0.52) (3.65) (0.60) (3.48) Dummy for specialized education 1.60* 0.92 1.48* 0.09 (0.56) (3.30) (0.64) (3.10) Dummy for higher education 3.40* -3.74 3.40* −5.89 (0.81) (4.76) (0.90) (4.87) Dummy for administrator −0.22 −3.23 −0.93 −2.52 (0.75) (3.83) (0.81) (3.57) Years of tenure 0.051 1.11* 0.18* 1.16* (0.089) (0.26) (0.087) (0.25) Dummy for at 0.13 n.u. 1.27 n.u. least one dependent (0.73) (0.73) Dummy for no one else works −0.25 n.u. −0.56 n.u. (0.60) (0.68) Net reduction 1991-92 0.26* n.u. 0.23* n.u. (0.37) (0.040) Net reduction 1991-93 0.003* n.u. −0.0045* n.u. (0.0009) (0.001) Net reduction 1991-94 −0.26* n.u. −0.23* n.u. (0.04) (0.041) Outside earnings −0.103 n.u. −0.142 n.u. (0.065) (0.059) Number of dependents n.u. 4.13* n.u. 3.94* (1.1) (0.96) Number of children n.u. −2.29 n.u. −2.97 (1.53) (1.52) Number of adults n.u. 2.10 n.u. 1.90 (1.21) (1.18) Number of pensioners n.u. −0.12 n.u. 0.26 (1.41) (1.42) Number of students n.u. −2.96 n.u. −4.13 (2.23) (2.15)
OCR for page 195
Transforming Post-Communist Political Economies 1. Only Nonseparated 2. Nonseparated and Unemployed Inside Earnings Outside Earnings Inside Earnings Outside Earnings Number who work n.u. 2.17 n.u. 4.39* (1.42) (1.46) Inside earnings n.u. −0.43 n.u. −0.16 (0.74) (0.77) Observations 1,191 1,191 1,321 1,321 NOTES: Standard errors are in parentheses. Coefficients significant at the 5 percent level are indicated with an asterisk (*). "n.u." denotes that this variable was not used. the last two also include the unemployed. In none of the specifications do we find these measures of inside and outside earnings to have significant negative signs. Again, however, this is not necessarily cause for rejecting Hypothesis I because the cross-tabulation evidence clearly indicates that most people obtained a relatively high level of outside earnings. Relevant for Hypothesis 2, the dummies for women and for pensioners have a significant negative sign in the inside earnings regression, but are not significant in the outside earnings equation. The education dummies are significant in the inside regression, but also not significant in the outside equation. By this measure, it seems that no one was unable to compensate for poor low inside earnings. Women earned an average of $28.9 from survival strategies, while men earned $29. Pensioners were somewhat behind, with an average of only $22.5, but their formal-sector earnings were a couple of dollars higher than those of nonpensioners. COPING AND NOT COPING We consider a person to have been coping if his or her monthly income, including all survival strategies, was higher than $25 per month. This is a crude but probably robust measure. A person is considered not to have been coping if his or her total monthly income was less than $15. In our sample, 56 percent of men and 44 percent of women met this definition of coping, while 20 percent of men and 26 percent of women did not.19 Table 7-7 shows the results for probit regressions. In the first column, the dependent variable was equal to 1 if total income exceeded the $25 coping 19 These definitions are of course fairly arbitrary, but they do capture the distribution of incomes.
OCR for page 196
Transforming Post-Communist Political Economies TABLE 7-7 Probit Regressions for Coping and Not Coping Left-Hand-Side Variable Coping (income over $25) Not Coping (income under $15) Constant −1.60* −0.064 (0.19) (0.19) 18-25 years old −0.44* 0.35 (0.21) (0.21) 26-35 years old 0.084 −0.075 (0.17) (0.18) 36-45 years old 0.42* −0.38 (0.16) (0.18) Female dummy −0.91 * 0.44* (0.13) (0.13) Dummy for specialized education 0.46* −0.033 (0.14) (0.14) Dummy for higher education 0.78* −0.070 (0.18) (0.19) Years of tenure −0.015 0.0044 (0.013) (0.013) Number of dependents −0.00088 0.00073 (0.0053) (0.0042) Outside intensity 0.15* −0.227* (0.009) (0.023) Inside intensity 0.0031* −0.0040* (0.00044) (0.00046) Dummy for unemployed with assistance 0.51 −0.25 (0.53) (0.63) Dummy for unemployed without assistance 0.49 0.060 (0.60) (0.78) Observations 1,451 1,451 level, and in the second column it was equal to 1 if total income was less than the $15 minimum coping threshold. The coping regression shows women were less likely to be coping in these ways, while those with specialized secondary or higher education were more likely to be coping. In confirmation of Hypothesis 3, we find that both inside and outside work intensity are significantly positive. The second column of Table 7-7 is the noncoping regression, and it again shows the female dummy to be significant, even when we control for the inside and outside intensity of work. Although not shown here, the dummy for pensionable age is not significant in any specification for either regression. The likely explanation is that women had less access to some remunerative strategies. In particular, women did not work as taxi drivers because they did not own cars and did not generally work as drivers in the formal sector. Women probably took fewer trips (except to Russia) because of the risks involved. They also did less "parallel work" for some reason, possibly be-
OCR for page 197
Transforming Post-Communist Political Economies cause of discrimination. It could also be that the nature of work in the nonstate sector does not favor women, but we are skeptical of this explanation because women previously did all kinds of work in the Soviet Union. At the same time, our results raise a question about an important hypothesis of the established literature. Unemployment as a dummy is significant, but not when we control for the outside intensity of work. Undoubtedly, this is because of multicollinearity because the outside intensity is necessarily zero for unemployed people. However, it also suggests that access to survival strategies is at least as important as formal employment status in determining whether people are able to cope on their own. CONCLUSIONS AND POLICY IMPLICATIONS How well does the established theory, based on Eastern European experience, explain labor market adjustment in Ukraine? It completely ignores informal activities, while the evidence indicates this is an important part of both how people survive and how they behave in the formal labor market. People can stay in the state sector and supplement their incomes with informal activities. Increases in outside "survival" work appear to be complementary to reductions in inside "formal" work. We find that the use of survival strategies is an important determinant of whether people can or cannot cope on their own. At least in Ukraine, there has been a considerable flow of people between formal jobs, but unemployment remains low. The established theory of post-communist labor markets needs to be modified to include self-employment and the ability of people to generate their own incomes (albeit at a low level). The Ukrainian nonstate sector, mostly individuals operating by themselves and in an informal way, expanded massively during 1992-1994. This growth occurred in large part because the collapse of the state sector and increasing administrative controls over the official economy meant people had to find ways of supplementing their official incomes. At the same time there has been de facto liberalization of entry into informal activities. A substantial amount of income-earning assets is already controlled by private individuals. Real estate, vehicles, and some agricultural land are already in private hands to a significant degree (even if not formally privatized). State assets such as trucks, trains, and industrial premises can be accessed easily through informal markets. It is fortunate that the state lost its ability to prevent people from using a wide range of assets as they see fit, because this enabled them to earn higher incomes. However, without proper privatization there is insufficient investment in these assets. Ukrainian people survived on their individual and collective inherited capital stock, and unless full private ownership is established, this capital will likely not be replaced as it depreciates.
OCR for page 198
Transforming Post-Communist Political Economies It is important to stress, however, that until mid-1994, the Ukrainian private sector merely offered a way to survive. There was not enough stability for serious private-sector investment to take place. People working in this sector were earning just enough to survive. The private sector can be a powerful force for the positive transformation of work after communism. The clearest example of this effect is in Poland, where almost every aspect of the economy has been fundamentally transformed as a result of the emergence and growth of new business over the past five years (Johnson and Loveman, 1995). But this form of private-sector development requires an environment in which private business can establish itself in activities that require relatively sophisticated organizations. This happens only when private entrepreneurs are willing and able to make significant investments in fixed capital assets. Preliminary evidence suggests that the stabilization and proper liberalization pursued since the fall of 1994 have begun to have these effects. ANNEX 7-120 UNEMPLOYMENT Our theory and evidence suggest a reinterpretation of what has happened to unemployment in Ukraine. Officially registered unemployment in the summer of 1994 was below 0.5 percent. This figure is remarkably low and much lower than the rate in most post-communist countries of Eastern Europe, where it ranged between 10 and 20 percent. It is commonly supposed that the official numbers are understatements, but even surveys indicate that open unemployment—people who do not have a job and would like one—in Ukraine is only between 3 and 6 percent. The established explanations for the relatively low level of unemployment in Ukraine are that very few people are fired from state enterprises, and that there are many genuine unemployed who do not register because benefits are so low (see Coricelli and Revenga, 1995). We examine these hypotheses in turn. Firing and Hiring At the beginning of 1991, the organizations in our sample had total employment of 13,498, and this figure remained approximately constant in 1991 and 1992; there was actually a 1 percent increase, most of which occurred in 1992. Of 26 organizations, 16 had an increase in employment during this period, and there were increases in all sectors, although a higher proportion of firms increased employment in the machine- and equipment-building sectors. However, total employment fell 5 percent in 1993, and a further 5 percent from the start of 1994 to the time of our interviews in mid-1994. At the time 20 An annex containing a detailed analysis on "gross separations" is available on request.
OCR for page 199
Transforming Post-Communist Political Economies of the interviews, our sample therefore shows a total net decline in employment of 9 percent from January 1991 and a total net decline of 10 percent from January 1993. From January 1991 to the time of the interviews, the employment of women fell by 10 percent and that of men by 5 percent. Total employment in the sampled organizations at the time of our interviews was 11,664. This net decline in employment masks the fact that there was a substantial amount of hiring in these same enterprises during 1992-1994. Strikingly, 24 out of 25 enterprises for which we have this information hired some people, and while only 3 enterprises had gross hiring above 10 percent of their January 1992 employment levels, 17 hired more than 5 percent of this total.21 In sum, the evidence indicates that people have been fired from all the state organizations we surveyed. It is true that firings were minimal until January 1993. As part of a labor force reduction, only 0.5 percent of men were fired during 1991 and 0.3 percent in 1992, although 1 percent of women were fired in both years. However, in both years, "other" firings were between 1 and 2 percent; this figure probably represents managers getting rid of workers they did not want. After January 1993 there was a sharp increase in firing: 3 percent during 1993 and 5 percent from the beginning of 1994 as labor force reduction, and 4-5 percent as "other." Even though Ukraine has lagged in terms of overall economic reform, there has been a substantial labor force reduction in many state enterprises. Many people have either been fired or been forced to quit. Why, then, has unemployment remained so low? The Unemployed There were 351 people who identified themselves as unemployed among our respondents. Of this total, 19 percent were not receiving any benefits and were not registered as being unemployed, and 38 percent had been unemployed for less than 3 months. For up to 3 months from separation, people who have been fired as part of a labor force reduction get their benefits from the enterprise rather than the government. According to Ukrainian practice, people are not counted as unemployed until they register with the government's offices. The effect of this procedure is to exclude almost all short-term unemployed from the registered total.22 21 According to Ukrainian law, a worker can be dismissed as part of a labor force reduction only if he or she cannot or will not take another job in the enterprise. The simultaneous firing and hiring on the scale reported in our sample suggests that this legal requirement is not enforced. 22 In the case of dismissal due to a labor force reduction, management must give the employee at least 2 months' prior notice. On the day of dismissal, the employee must receive any outstanding payments, including compensation for unused holidays, and an additional I month's pay. During the next 2 months, any ex-employee who has not found another job has the right to
OCR for page 200
Transforming Post-Communist Political Economies The reemployment rate for quits is high. Annex Table 7A-1 shows that almost all quits among those surveyed were reemployed after 1 month and that the reemployment rate for people who had been fired rose steadily. Almost no one who had quit had registered as being unemployed or was receiving benefits.23 Therefore, in our survey, the registered unemployed (reported in national statistics) were only those fired people who had been unemployed for at least 3 months. This is just 34 percent of all the unemployed people in our sample, and converts to an unemployment rate among the population of 4 percent. If we assume that only people unemployed at least 4 months were registered unemployed, the implied unemployment rate falls to 2 percent.24 The reemployment rate among our sample was very high for anyone who would have been eligible for benefits. The average reemployment rate for fired people seems rather low at 28 percent, but it rises to 47 percent for people unemployed 4 or more months and keeps on rising (Annex Table 7A- 1). It appears that people were not particularly attracted by the level of unemployment benefits. At least in our sample, there was almost no long-term unemployment. Although our results imply a registered unemployment rate of 2-4 percent, the "true" unemployment rate—people without a job who are looking for work—in our sample is 11 percent (including both quits and people who had been fired).25 If we assume that pensioners who had separated in the last 3 years would also have liked to work but could not, the unemployment rate increases to 14 percent. continue to receive a monthly salary from the enterprise. According to Ministry of Labor estimates, about 70 percent of employees who have been dismissed receive such allowances at their enterprises. To have a right to unemployment compensation, an employee must register at his/her local employment office during the 3 months following dismissal. If this is not done in time, the employee forgoes the right to unemployment compensation. 23 Table 7-1 shows that the unemployed without assistance (primarily quits) were heavily engaged in working at their dachas and in taking trips to Russia. These were unemployed people who were looking for work, and who had significant side incomes that they presumably did not want to report to the state. 24 This estimate is higher than the official statistic for a number of reasons. Our sample did not include young people who had never worked before or pensioners. Our sample also contained more industrial employment than is the case for the economy as a whole. About 20 percent of Ukraine's population lives in rural areas, closely linked to agriculture, and the unemployment rate is probably lower there. Our sample may also have been biased toward including more people who had separated recently, and this would also tend to increase the unemployment rate. 25 Taking gross separations since January 1993, removing retirements, and applying the differential reemployment rates for quits and fired implies that a total of 1,557 people were unemployed. This would be 11 percent of the January 1993 level of employment among the surveyed firms. We assume that everyone who had separated before January 1993 was reemployed. Given the high rates of reemployment for people separated more than 9 months, this is not unreasonable.
OCR for page 201
Transforming Post-Communist Political Economies TABLE 7A-1 Rate of Reemployment Number of Months Since Separation Reason for Separation Quit Fired 1 or more 96 34 2 or more 96 34 3 or more 99 39 4 or more 100 47 5 or more 100 54 6 or more 100 70 7 or more 100 82 8 or more 100 90 9 or more 100 91 Whole sample 69 28 In short, our work suggests that the actual unemployment rate in Ukraine is close to that in Eastern Europe. However, the peculiarities of the benefit system, with the enterprise paying for the first 3 months, mean that most of the short-term unemployed are not registered. Although this implies an overall unemployment rate similar to that in Eastern Europe, Ukraine has much less long-term unemployment. Most likely this is because the benefits are so low—typically around 20 percent of the minimum survival wage—while it is relatively easy to find a new job of some kind and to supplement that income with a survival strategy. In contrast to the established view, separated people in our sample could readily find new formal jobs, and labor force participation had not declined as of the summer of 1994. Almost everyone was reemployed within 6 months. Even more important, people could generate new informal income-earning opportunities for themselves. This meant people were more likely to quit voluntarily and create job openings, even in companies that were contracting. As a result, in the summer of 1994, very few people in Ukraine were not working.
OCR for page 202
Transforming Post-Communist Political Economies REFERENCES Aghion, P., and O. Blanchard 1994 On the speed of transition in Eastern Europe. Macroeconomics Annual. National Bureau of Economic Research. Alexeev, M., and V. Treml 1993 The Second Economy and the Destabilizing Effect of its Growth on the State Economy in the Soviet Union: 1965-1989. Berkeley-Duke Occasional Papers on the Second Economy in the USSR, No. 36, December. Åslund, A., P. Boone, and S. Johnson 1994 Ukraine: Ready for a breakthrough. Ostekonomisk Rapport 24:6(8). Blanchard, O., S. Commander, and F. Coricelli 1995 Unemployment and restructuring in Eastern Europe and Russia. In Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia, S. Commander and F. Coricelli, eds. Washington, DC: The World Bank. Commander, S., and F. Coricelli, eds. 1995 Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia. Washington, DC: The World Bank. Coricelli, F., and A. Revenga 1992 Wages and Unemployment in Poland: Recent Developments and Policy Issues. Policy Research Working Paper Series, WPS821, January, The World Bank. 1995 A puzzling job. The Economist February 18th. Friedman, E., and S. Johnson 1996 Complementarities and Optimal Reform. Unpublished paper, November, Duke University. 1997 Dynamic monotonicity and comparative statics for real options. Journal of Economic Theory 75(1):104-121. Johnson, S., and G. Loveman 1995 Starting Over in Eastern Europe: Entrepreneurship and Economic Renewal. Boston: Harvard Business School Press. Johnson, S., and O. Ustenko 1995 The Road to Hyperinflation: Ukraine 1991-93. In Economic Strategy in Newly Independent Post-Communist Countries, M. Wyzan, ed. New York: Praeger. Kaufmann, D. 1994 Diminishing returns to administrative controls and the emergence of the unofficial economy: A framework of analysis and application to Ukraine. Economic Policy 19 (December). Layard, R. 1994 Introduction. Russian Economic Trends (3)2. Leitzel, J. 1995 Russian Economic Reform. London: Routledge. Milgrom, P., and J. Roberts 1990 Rationalizability, learning, and equilibrium in games with strategic complementarities. Econometrica 58(6): 1255-1277. 1994 Comparing equilibria. The American Economic Review 84(3):441-459. Standing, G. 1994 Results from the Labor Flexibility Study in Ukraine. ILO report. Zienchuk, M. 1994-1995 Ukraine in Numbers. Monthly Statistical Bulletin of the Ukrainian Ministry of Economics.
Representative terms from entire chapter: