ریسک قوانین و مقررات تجارت با تحرک نسبی نیروی کار

We examine the welfare consequences of terms-of-trade risk in a small open economy in which it is costly for workers to move between sectors. Relocation costs lead to partial labor mobility, sectoral wage gaps and income risk exceeding that of an economy in which relocation is costless. Using observed wage differentials and standard values for volatility and preferences, we find that the welfare cost of partial labor mobility alone is unlikely to be very large, even in the absence of self-insurance arrangements. In addition, modest consumption substitution elasticities significantly reduce these welfare costs.

Small open economies face large and unpredictable swings in their terms of trade. The ubiquity of these shocks and their implications for business cycles were first discussed by Mendoza (1995) who analyzed data on the G-7 industrial nations and 23 developing countries. In his sample, the standard deviation of terms-of-trade shocks was 4.7% for the G-7, and about 12% for the developing countries. More recently, using a sample of 66 developing countries, Bidarkota and Crucini (2000) find that among the top quartile of countries exhibiting the highest terms-of-trade volatility, the standard deviation of the country terms of trade was an extraordinary 25% per year. For the next three quartiles, the standard deviations were 16%, 12.5%, and 8.5%, respectively. The substantial income risk induced by terms-of-trade shocks is therefore of considerable concern for a wide range of economies. In part, the burden of these shocks can be cushioned by reallocating expenditures across sectors, so as to minimize consumption volatility. Moreover, on the production side, when agents respond to reallocation incentives instantaneously and costlessly, it is only the aggregate consequences of terms-of-trade shocks that matter given that marginal factor products are always instantly equalized across sectors. However, the assumption of dynamic and reversible short-run resource reallocation may significantly understate the welfare effects of terms-of-trade volatility. Short-run adjustment decisions involve a balance of the costs and benefits of relocating, which may lead agents to approach resource reallocation in a seemingly “sluggish” manner. In recognition of these costs, the theory of international trade has traditionally incorporated a “fixity” of factors of production, particularly in specific factor models. When agents respond sluggishly to changes in incentives, sectoral marginal products will not be equalized and some sectors of the economy will be disproportionately affected by a given external shock. This raises the possibility of additional domestic and sector-specific risks, which may compound the welfare costs of external shocks. How important are the economic and welfare consequences of costly resource reallocation in an economy exposed to substantial terms-of-trade volatility? To address this issue, we analyze consumption risk within a two-sector dynamic general-equilibrium framework where workers must incur a fixed cost before relocating to another sector, thereby introducing endogenous partial labor mobility. As forcefully argued by Dixit and Pindyck (1996), among others, any attempt to model the cost of frictions must take into account the recurrent nature of relative price changes and the consequent general equilibrium effects on the distribution of resources in the economy, as these elements can lead to sharply different conclusions about welfare. We therefore include both of these elements in our framework. Because workers in the traded and non-traded goods sectors face different wage prospects, we are simultaneously able to assess the welfare costs of overall labor immobility as well as potentially uninsurable sector-specific labor-income risk. Using empirically available sectoral wage differentials and parameter values for terms-of-trade volatility and preferences taken from the literature, our results suggest that the welfare cost of partial labor mobility alone is unlikely to be very large, even in the absence of self-insurance arrangements. Specifically, we find that the welfare cost of partial labor mobility is typically less than 2% of lifetime consumption, and about two-thirds to half of this cost is attributable to sector-specific labor-income risk that can be diversified away by domestic risk-sharing arrangements.1 However, the relative welfare significance of partial immobility depends on a balance of the frictions that impede labor reallocation and the substitution effects that cushion terms-of-trade shocks. In particular, economies with a low elasticity of substitution between traded and non-traded goods and between domestic and foreign traded goods exhibit considerably larger welfare losses for any given fixed cost of relocation and terms-of-trade volatility. Our analysis is related to the recent international risk-sharing literature which, using the techniques proposed by Lucas (1987), quantifies the welfare cost of consumption variability in open economies. Because the benefits of risk sharing are proportional to the amount of diversifiable risk, this literature provides an estimate of the welfare costs of consumption volatility. For instance, using a business-cycle model, Mendoza (1995) provides one of the first attempts to quantify the welfare costs of consumption volatility attributable to the terms of trade. Quite often, however, explicit links between consumption volatility and terms-of-trade fluctuations are not made. For example, several studies have treated consumption as an endowment process (Cole and Obstfeld, 1991 and van Wincoop, 1999), focused on productivity shocks (Tesar, 1995), or simply assumed that the variance of consumption growth is proportional to the variance of the terms of trade (Bidarkota and Crucini, 2000). Our approach ties consumption risk directly to terms-of-trade induced income volatility in a small open economy. To derive the variability in consumption as a function of the variability of the terms of trade, we explicitly model employment choices and the production side of the economy. We then compute the risk posed by term-of-trade fluctuations. We also extend this literature by incorporating domestic sector-specific risk. In their discussion of the literature on the welfare benefits of risk-sharing, Obstfeld and Rogoff (1996, p. 331–332) call for the “explicit modeling of consumer heterogeneity” to assess the welfare costs of uninsurable labor-income risk. In our model, heterogeneity and sector-specific labor-income risk arise endogenously through the interaction between costly labor relocation and fluctuations in the terms of trade. Thus, our main contribution to this literature consists of modeling endogenously determined imperfect labor mobility, and computing the size of the welfare costs of terms-of-trade fluctuations resulting from partial labor mobility. The consequences of the frictions that we introduce are important if not dominant features of the economic landscape. Many attempts to stabilize commodity prices or farm incomes have been based on the conviction that there is imperfect sectoral factor reallocation and the resulting sector-specific labor-income risks are uninsurable. Indeed, the debate over external shocks and trade liberalization has drawn attention to the costs of adjustment and their redistributive and reallocative consequences.2 For a recent example, concern over sector-specific welfare costs related to “offshoring”—the loss of white-collar jobs overseas—is motivating attempts in the U.S. to extend Trade Adjustment Assistance (TAA) beyond manufacturing to service sector workers as well.3 The rest of the paper is structured as follows. Section 2 lays out our basic model. Section 3 describes how we measure the risk caused by terms-of-trade variability (both with and without perfect labor mobility) and presents our results. Section 4 concludes.

Using preference and terms-of-trade volatility parameters taken from the literature, we find that the total welfare costs of partial labor mobility are economically small even in the absence of any form of precautionary saving to smooth consumption over time. The model we consider rules out divergent consumption paths across workers employed in different sectors within a country, limiting both the potential gains to domestic risk sharing and the aggregate effects of partial labor mobility. Wage differentials may follow a random walk within the zone of inaction, but thresholds act as reflecting barriers that prevent these differentials from growing too large, thereby endogenously limiting the welfare costs (for plausible parameterizations of the elasticity of substitution between traded and non-traded goods). To place these results in context, note that the substantial benefits of more extensive international risk sharing arise only when one allows for divergent consumption paths across countries (van Wincoop, 1999). One important aspect of our modeling strategy is the absence of any capital or storable good. This clearly overstates the potential welfare costs of labor immobility because (i) trade theory suggests that even a single mobile factor of production can substantially help equalize returns across sectors, and (ii) international asset trade and precautionary savings can cushion the welfare consequences of fluctuations in income. Our choice of financial “autarky” is intentional, however, and is a useful starting point: failure to demonstrate substantial welfare costs even in the absence of asset markets and self-insurance indicates limited benefits from risk-sharing arrangements (including price stabilization programs). A qualification to our conclusions is that we use a simplified labor market with fixed labor supply decisions and full employment. Both of these elements have potentially important implications for real wages and welfare. While being able to vary labor supply can help workers mitigate sector-specific and aggregate risks, unemployment may accentuate these risks, and it remains an open issue whether the balance of these mechanisms can significantly alter our findings.