Economic Origins of Dictatorship and Democracy

by Daron Acemoglu

  Globalization has the opposite effect on factor prices in rich countries. Whereas wages increase and the returns to capital fall in the labor-abundant developing nations, wages should fall and returns to capital should increase in the capital-rich nations, such as the OECD economies. Should we expect a greater likelihood of coups against democracy in the OECD countries? We believe the answer is no: because the OECD societies are already fully consolidated democracies, a marginal increase in democratic redistribution will not push them into the position of unconsolidated democracies.

  This discussion is predicated on the presumption that increased international trade reduces inequality in developing countries, especially narrowing the gap between capital holders and labor. This is a prediction of most trade models (especially the celebrated Heckscher-Ohlin model) when the nondemocratic countries are scarce in physical capital and abundant in labor relative to the rest of the world. But, in practice, some of the nondemocratic countries joining the world economy may be abundant in land (e.g., Argentina and Chile in the early twentieth century). In this case, international trade may increase the return to land and, through the mechanisms emphasized in Chapter 9, make democratization and democratic consolidation less likely.

  Another important caveat is that international trade not only affects the relative price of capital and labor but also the relative price of human capital (i.e., returns to skills). Less developed nations are typically scarce in skilled labor, and we should expect increased trade integration to reduce the skill premium in those countries. However, recent experience in many of those countries has been an increase in the returns to skill and a greater gap between the more and the less educated workers. The literature in economics explains this fact by the associated diffusion of skill-biased technologies to less developed nations, increasing the marginal product of skilled workers. Therefore, we also briefly discuss a model in which trade integration increases the returns to skills and show that even though this may increase overall inequality, it might again help democratic consolidation. We can think of the skilled workers as corresponding to the “middle class” and, as discussed in Chapter 8, with a richer middle class, democracy is less redistributive and may be more likely to consolidate.

  Another aspect of globalization is increased financial integration. We also show in this chapter that increased financial integration in the world economy may affect the creation and survival of democracy. For this purpose, it is important to distinguish between the effects of capital inflows to less developed and capital-scarce nations, which follows increased financial integration and the possibility of capital flight from those nations. Capital inflows, just like increased international trade, reduce returns to capital in the capital-scarce countries and increase wages. Therefore, the effects of financial integration through this channel are similar to those of increased international trade: they reduce the income gap between the rich and the poor and, by this channel, may help the consolidation and creation of democracy.

  The possibility of capital flight may also help democracy but this time through a different mechanism. In a closed economy, without the possibility that capital may be invested abroad, the elasticity of capital supply is relatively low: capital holders can consume their capital rather than invest it or perhaps invest it in nontaxable activities (in the “informal sector”), which may have much lower productivity. In contrast, with increased financial integration, high taxes may encourage capital holders to take their capital out of the country and invest it in other markets where taxes are lower and where the risk of expropriation is absent. In other words, financial integration increases the elasticity of capital supply. Realizing this, democracy imposes lower taxes on capital and is generally forced to be less redistributive. Therefore, financial integration also reduces redistribution in democracy because of the potential flight of capital. Anticipating this, the rich have less to fear from democracy, are more willing to accept it rather than use repression, and are less willing to undertake a coup against democracy when a window of opportunity arises. As a result, financial integration may also help the creation and consolidation of democracy.

  Lastly, globalization also comes with increased political integration, which may affect the costs of coups through various channels. Most important, with increased political integration, countries may face greater sanctions from other democratic nations if they suffer a coup against democracy. Through this channel, globalization might help democratic consolidation. In addition, greater political integration may also help the development of civil society in less developed nations and increase the cost of coups and contribute to democratic consolidation.

  Overall, our analysis in this chapter reveals that there may be important links between globalization and the emergence and consolidation of democracy. An interesting possibility is that these links may help to explain waves of democratizations or coups. By a wave, we mean a concurrent move toward or away from democracy in a number of nations. Waves toward democracy may have occurred historically - for example, in the period before the First World War, after the Second World War, and since the 1970s (Huntington 1991; Markoff 1996). It is interesting that there is a close correlation between these waves of democratization and upsurges in globalization; indeed, it is natural to think of globalization as being a phenomenon simultaneously impacting many nations and thus as a potential explanatory variable.

  As noted previously, many scholars have discussed the relationship between globalization and democracy and our analysis makes several important contributions. First, to our knowledge, no one has previously suggested that increased international trade can influence the creation or consolidation of democracy through the channels we discuss (namely, the impact on factor prices and, hence, the distribution of income).

  Second, Bates and Lien (1985), Bates (1991), Rogowski (1998), Newman and Robinson (2002), and Boix (2003) note that the possibility of exit from a nation might promote democracy, but they do not offer an analysis of the full political equilibrium when international trade affects the structure of inequality and the options of various parties in the political game.

  Third, by placing the idea of exit into a standard economic model of factor mobility, we discover other important effects — for example, the distributional impacts of capital inflow. Although there is a huge literature in economics on the impacts of globalization, both increased trade and financial integration (Prasad et al. 2002), it has only just begun to link these forces to institutional change. Scholars such as Rodrik (1997) and Garrett (1998) emphasized the idea that increasing globalization limits the policy scope for national governments, but they do not suggest that this may influence the equilibrium structure of institutions. We show that globalization may have important effects for democracy.

  Fourth, most of the literature in political science has focused on ideas about geopolitics and the ideological diffusion of democracy (Kopstein and Reilly 2000; Maxfield 2000).

  The results presented in this chapter are suggestive but have yet to be tested empirically. Moreover, whether the mechanisms we discuss promote democracy depends on which part of the parameter space we are. This was discussed previously with respect to the effects of increased trade integration on inequality, and the same applies to the effect of financial integration. We see it as a theoretical possibility, although probably not the empirically relevant case, that by reducing inequality, increased trade integration may consolidate nondemocratic regimes.

  Finally, globalization may reduce the scope for democracy to set majoritarian policies by so much that the creation of democracy fails to promote stability. If democracy delivers nothing to the citizens, then revolution becomes attractive for them and repression becomes attractive for the elites. In such a circumstance, globalization does not promote democracy.

  2. A Model of an Open Economy

  To study the links between globalization and democracy, we use a version of the model from Chapter 9 with capital, land, and labor. We focus on a sing
le country, which is first taken to be closed to international trade. Then we look at the case in which the country integrates into the world economy and starts trading goods with other countries.

  As in Chapter 9, we assume there is an aggregate production function but instead of this directly taking capital, land, and labor as inputs, we assume that three different intermediate goods are used as inputs. As before, let Y be the output of the final good that is consumed and let YK, YL, and YN be the amounts of the three intermediate goods used in the production of Y. The aggregate production function is again assumed to be Cobb-Douglas:

  (10.1)

  and we assume, as before, that 0 < θ < 1 and σ > 0.

  In a closed economy with no trade, intermediate goods are themselves produced by domestic factors of production; the subscripts on the intermediate goods indicate that one is capital-intensive, YK; another is land-intensive, YL; and the last is labor-intensive, YN. In an open economy, intermediate goods are traded internationally. On the production side, in the simplest possible world, all three goods are produced using only their respective factors; therefore, domestic production of each intermediate good is given by:

  (10.2)

  exploiting the fact that there are 1 — δ workers. The remaining δ agents, who constitute the elites, do not own any labor, and each holds fractions δ of the total capital stock, K, and the total land stock, L.

  When there is no international trade, this world is identical to the one in Chapter 9. More formally, without international trade, the country in question has to use its domestic production of capital, land, and labor-intensive intermediate goods to produce output. Substituting (10.2) into (10.1), we have that:

  which is identical to the aggregate production function specified in Chapter 9.

  We assume that all markets, both for intermediate inputs and factors of production, are perfectly competitive. We set the price of final output to be 1 and use this good as the numeraire. The prices for the three intermediate goods are denoted by PK, PL, and pN. To determine these prices, we examine the cost-minimization problem of a firm choosing input demands to minimize the cost of production. Formally, a firm solves the problem:

  subject to:

  Here, pKYK + pLYL + pNYN is the total cost of using the three intermediate goods. This is a simple constrained-optimization problem. To solve it, we form the Lagrangean function:

  and derive the first-order conditions with respect to the three choice variables YK, YL, and YN. These are:

  (10.3)

  From these, we derive:

  (10.4)

  where the first follows from dividing the first and third equations in (10.3), and the second follows from dividing the first two equations in (10.3). These equations imply that:

  (10.5)

  Because one unit of each factor is used to produce its respective goods and factor markets are competitive, each factor is paid the value of its marginal product - that is:

  where w denotes the wage rate, r the return to capital, and ν the rental rate of land. In the closed economy, we use (10.2) together with (10.5) to obtain:

  (10.6)

  and, therefore, the relevant closed-economy factor prices are identical to those in Chapter 9 and given in (9.2). This is, of course, not surprising given the previous observation that the two models are identical for the case of the closed economy. Consequently, factor shares are also the same as before and given by (9.3). Exactly as before, we have that incomes of the poor citizens and of the elites and average incomes are given by (9.4), (9.5), and (9.6). We again make an assumption ensuring that the elites are richer than average (i.e., θ > δ). Thus, the elites are homogeneous and each owns capital and land in equal amounts. The citizens simply own their own labor. We focus in this chapter on situations in which political conflict is along the lines of socioeconomic class. Nevertheless, as is clear from our previous analysis, many of the results apply when political conflict is along other lines; we return briefly to this issue in the conclusion of this chapter.

  We again assume there is a single tax rate on income, irrespective of its source. This tax rate creates the standard distortions captured by the function C(τ). Then, the most preferred tax rate by a citizen, τP, is given by an equation identical to before.

  2.1 Factor Prices and Incomes in an Open Economy

  Now assume that this country joins the world trading system and can trade with all other countries in the world without any friction. We think of this increased trade integration as one aspect of “globalization.” Because there is only a single produced good, there is no incentive for countries to trade it. However, there may be incentives for countries to trade the intermediate goods that are inputs to the production of the final good because they are produced using factors of production with which countries may be differentially endowed. For now, we assume that factors of production cannot be traded (i.e., there is no capital mobility and no migration). Instead of simply using domestic stocks of capital, land, and labor to produce intermediate goods, a country can trade with the rest of the world, sell its supplies of capital-, land-, and labor-intensive intermediate goods at world prices, and attain a different level of production of the final good. If world prices of capital-, land-, and labor-intensive goods are denoted, respectively, by PN’pK’ and pL (underlined variables always refer to the open economy), then the budget constraint of this country is:

  The left-hand side of the equation is the total expenditure of this country on intermediate goods at world prices; the right-hand side is the total revenue that this country raises by selling its production of intermediate goods at world prices (considering the production functions given by (10.2)).

  How are these world prices determined? The answer is not essential for this discussion. We simply assume that the prices are determined in some world-market equilibrium, and we take it such that:

  (10.7)

  where we can think of Ψ as the ratio of the sum of capital and land relative to labor in the world economy. In (10.6), what mattered for the determination of domestic prices was the ratio of K + σ L to 1 — δ; this is what is meant by “the ratio of the sum of capital and land relative to labor.” In the world economy, we can think of the same ratio mattering but where the relevant totals are the world stocks of factors, not just the stocks in one country. For example, if all countries of the world trade and there are no tariffs or trading frictions, we have that:

  where Nj is total labor supply in country j, Kj is the capital stock, and Lj is the stock of land. In this case, we have that ψ is equal to the sum of the capital-and-land-to-labor ratios across the world — i.e.,

  If, on the other hand, there are tariffs or trading frictions, ψ differs from this ratio. Whether this is the case is not central for the analysis in this chapter.

  Our focus is with the emergence and consolidation of democracy in nondemocratic societies. Nondemocratic societies are typically poorer and, therefore, they are more abundant in labor than capital. Therefore, it is natural to think that the country in question is relatively scarce in capital. Stating this as an assumption, we have:

  (10.8)

  The most important implication of this assumption is seen by comparing (10.7) with (10.6), which implies that after trade opening, the price of the labor-intensive intermediate good increases in the country in question (which is presumed throughout to be a relatively labor-abundant country). Intuitively, this country is relatively abundant in labor compared to the world economy, which depresses the price of the labor-intensive intermediate good when there is no international trade. International trade pulls the price of the labor-intensive good to the world level.

  Once these prices are given, factor rewards in this economy are again given by the relevant value of marginal products, now evaluated at these world prices; therefore:

  (10.9)

  This implies that international trade also increases wages relative to capital and land returns. These changes in relative factor prices are the main c
hannel by which international trade has an impact on whether democracy emerges or consolidates.

  It is also noteworthy that we are implicitly imposing factor price equalization. That is, with ψ interpreted as the world capital-and-land-to-labor ratio and PN, pK, and pL as the world prices, factor prices given by (9.2) would be the same as factor prices in other countries. Nevertheless, whether factor-price equalization holds is also not important for this analysis. What matters is simply that after trade opening, the price of the abundant factor increases relative to other factor prices and that for the country in question, a relatively poor country, the abundant factor is labor. This is a feature of many trade models even when factor-price equalization does not hold, as well as the standard Heckscher-Ohlin model with factor-price equalization. The important implication of this change in factor prices is that inequality declines after trade. Existing evidence is broadly consistent with the notion that wages are higher in capital-abundant countries (Trefler 1985; Leamer 1998; Romalis 2004), but there is also evidence that the recent increase in international trade has raised the incomes of higher skilled workers more; we discuss these implications next.