Comment and Discussion

Carmen Reinhart: I will offer three areas of comment on the paper. The first concerns the macroeconomic environment in which the transitions discussed take place. Second, I want to focus on the methodology employed and suggest some directions where the authors might develop some interesting insights by further parsing the data. Finally, I have a few reservations concerning the quality of the data.

The paper reviews evidence on labor market mobility in nine countries. Three countries (Argentina, Mexico, and Venezuela) come from Latin America, while the remainder (Albania, Georgia, Hungary, Poland, Russia, and the Ukraine) are transition economies in Eastern Europe. The period of study ranges from as little as two years (Georgia and the Ukraine) to eleven years (Mexico). The paper uses longitudinal labor force survey data to construct a transition matrix for each country, with each cell representing the probability of having labor force status j in period t + k conditional on having status i in period t. The authors consider six possible labor force outcomes (an issue I will return to later): out of the labor force, unemployed, formal salaried worker, informal salaried worker, self-employed, and farmer.

As a macroeconomist, I tend to think about the big picture when viewing issues of labor mobility. Table 1 in the paper and the subsequent discussion highlights the high level of macrovolatility in the countries studied. Many of the countries experienced large output swings during the sample period, and in others (Hungary and Russia) the period of analysis immediately followed a severe recession. I would additionally emphasize that other types of volatility, such as relative price movements between the traded and nontraded sector, can have important influences on labor market options beyond their effect on output.

Economic crises serve to highlight the effects of the kinds of macroeconomic volatility I have in mind. The sample includes episodes of currency, banking, debt, and inflation crises, so consideration of crisis dynamics is nontrivial. As an example, currency crises imply large changes in the real exchange rate, which can have a particularly adverse effect on the nontraded sector. If, say, a currency crisis precedes the collapse of the real estate and construction industries, we would likely see a large movement of the displaced workers from the formal to the informal sector. The authors should control for such macroeconomic effects in their analysis, and they might generate some useful observations by examining in detail certain crisis subperiods.¹

My second suggestion involves looking more closely within the labor market movements reported. I again come back to big picture economic currents, here financial liberalization and privatization, which were prominent in all of the countries in the sample. I would like to see the authors expand their matrix to make it possible to study movements within a given labor market sector that could shed light on the effects of these big picture policies.

Financial liberalization has had a profound impact on the structure of the economies considered. I recall traveling in Indonesia in 1995, when overnight it seemed the country went from five banks to one hundred and five banks. Financial liberalization likely played an especially important role in the former Soviet states. As a result, a significant portion of these economies shifted from agriculture or manufacturing into finance. It would constitute a major contribution to document how the associated reallocation of resources affected labor market mobility. Importantly, this would require looking within the formal salaried sector to measure movement into financial firms.

The period of study also coincided with substantial privatizing of state-owned enterprises. Such privatizations occurred in all of the transition economies, as well as in Mexico and Argentina. A common feature of these privatizations was subsequent downsizing—newly private firms laying off surplus labor. I would like to know the extent to which transitions out of the formal salaried sector can be related to firms' privatizing. In Argentina, for example, privatization has been associated with rising macro-unemployment. Does this relationship hold at the microlevel, and how many of the laid-off workers wind up taking jobs in the informal sector? Note again that these questions require looking within a given cell of...