The study answers five important questions about the effect of economic transition on occupational movements and the costs of such movement for individual workers. First, does the restructuring process lead to a higher rate of occupational reallocation? Both gross and adjusted measures show an unambiguous increase in occupational transitions after 1991, when reforms began. The number of people moving to another occupation was considerably higher during the first four years of reform (1991–95) than during the preceding six years (1985–91). During seven years of transition (1991–98), 42 percent of employed respondents changed their occupation—nearly twice as many as did so during the six pre-reform years. Occupational flows were most intense during the first five years of reforms; after 1996 the rate of occupational mobility began to fall. This decline could be partially due to the diminishing rate of structural change and the relative stabilization of occupational composition.

Second, does the restructuring process change the structure and direction of occupational mobility? We hypothesize that structural changes make occupational mobility more complex, in that changes in occupation, firm, and industry are occurring simultaneously. The data confirm this hypothesis, showing that the share of people who change occupation when changing firm or industry increased in 1991–98 compared with 1985–91 (with a decline in 1996–98).

A large increase in the magnitude of flows to service-providing occupations also occurred. Flows to these activities were almost three times higher in 1991–98 than during the pre-reform period. Comparing occupational structures in 1985 and 1998, we observe an increase in the share of managers, entrepreneurs, specialists in business and law, customer service clerks, people, and other service-providing workers (table 1). (Interestingly, the number of protective services workers rose by more than 200 percent for the same period). At the same time, there was a strong decline in the number of engineers and skilled laborers, possibly reflecting a shift away from goods-producing industries toward service-providing industries.

Third, what explains the increase in occupational mobility in transitional Russia? Are people involuntarily forced to change their occupation because previously accumulated skills become obsolete and unusable? Or does increased occupational mobility reflect the creation of new opportunities?

To answer these questions, the study examines whether occupational change is caused by a decline in returns to a previous occupation or by an increase in returns to alternative options. The analysis shows that the probability of occupational switching increases with returns to alternative occupations before and after the mobility decision. In other words, people respond to new opportunities and alternative options. People also respond to perceived future changes in payments, or returns, to their current occupation. However, a 10 percent decrease in returns to the current occupation brings about more occupational mobility than a 10 percent increase in returns to alternative occupations, suggesting that people are often forced to move to another occupation because their current occupation is losing its market value.

Fourth, what impact does the local economic environment have on occupational mobility? Are the magnitude and direction of occupational mobility different across Russian regions? We hypothesize that local labor market conditions—which reflect an uneven speed of structural change and unequal outside opportunities across regions—are critical determinants of occupational shifts.

Results of empirical analysis show that people change occupations more often in regions with a higher local job destruction rate and a higher employment share in newly created firms. However, a 10 percent increase in the local job destruction rate in the industrial sector resulted in more occupational changes than a 10 percent increase in the local share of employment in newly created firms. Again this result implies that a relatively large part of occupational mobility in the transition period is driven by destructive forces.

Among other local characteristics, the industrial employment concentration index is found to be negatively related to occupational mobility. This finding suggests that poor outside alternatives reduce the opportunities for occupational mobility.

Fifth, to what extent does occupational reallocation increase the discrepancy between previously accumulated human capital and the market demand for skills? Intuitively, we would expect the connection between previously acquired education and occupation in the new market economy to weaken as a result of a transitional shock. Changes in the demand for occupation-specific skills and overall changes in wage distributions across occupations may induce people to end their old careers regardless of how successful they were under the previous system and to begin a new career in a completely different field with lower skill requirements.

In fact, the data show that a significant number of people choose new occupations that do not correspond to their education. In 1998 only 38.9 percent of engineering graduates held jobs as professionals. Some had become managers or entrepreneurs, but some had accepted jobs as laborers or service workers with lower skill requirements.

Table 2 shows ranking indices and ranks of 28 occupational categories according to level of schooling and level of earnings in 1998. Although the correlation between the two ranking indices is high, some discrepancies are apparent. Entrepreneurs and farmers rank 1st in terms of earnings but 15th in terms of education. Physicists, mathematicians, and engineers rank fourth in terms of schooling but ninth in terms of earnings. Movement up the schooling ladder is thus not necessarily associated with upward earning mobility.

Movement to occupations that on average require fewer years of schooling (a switch from engineer or technician to salesperson or guard, for example) appears common during transition. The data show that downward occupational mobility is forced by negative demand shocks and poor firm performance.

Finally, was the change of occupation beneficial for individual workers? What are the returns to occupational mobility in terms of earnings and subsequent wage growth?

We test two conflicting hypotheses on the returns to occupational mobility. The "destructive" theory of occupational mobility suggests that people who are forced to move may lose benefits they had in their previous occupation. They may, for example, agree to lower wages than they had before their career change. In contrast, the "creative" theory of occupational mobility suggests that voluntary occupational switches in response to positive demand shifts and new opportunities bring additional benefits to individuals.

We test these hypotheses by estimating the returns to occupational mobility in terms of subsequent wage growth. The data show that occupational mobility reduces wage growth, thus supporting the "destructive" theory of occupational mobility during transition.

Table 1. Changes in Occupational Composition in the Russian Federation, 1985–98