Economic Origins of Dictatorship and Democracy
Page 47
Combining (10.7) and (10.9), we have that post-trade factor prices are given by:
(10.10)
Equation (10.8) implies that wages are higher and the returns to land and capital are lower than under autarky.
Using these factor prices, post-trade incomes are:
and
and average income in this case is:
(10.11)
Using these expressions, we obtain the most preferred (unconstrained) tax rate of the citizens as τP, which again satisfies the usual first-order condition:
(10.12)
and
where τP is the preferred tax rate of the citizens after trade, given by (10.12), and τP is their most preferred tax rate before trade. Thus, the citizens, whose income comes from supplying labor, prefer to set lower taxes after trade. This implies that after globalization, democracy becomes less redistributive because globalization reduces income inequality.
3. Political Conflict - Democratic Consolidation
We now incorporate this economic model into our political models. We begin with an analysis of democratic consolidation. The analysis mirrors those before, especially those in Chapter 9, Section 5, and the game tree in Figure 7.1 captures the strategic situation. We assume that a fraction ϕ of both capital and land is lost during a coup, so there are no differential costs depending on asset composition. We define two different coup constraints - one before and one after trade - and two threshold levels for ϕ, which we denote ϕ* and, for before and after trade.
This discussion implies that the values from democracy before trade are given by (9.7). Similarly, before trade, the values to citizens and to the elites following a coup are given by (9.24).
The coup constraint before trade, Vr(C, ϕ) > Vr(D), is identical to the one we derived before in (9.16) with ξ = 1:
(10.13)
After trade, the values from democracy change because of the changes in factor prices and are given by:
(10.14)
Similarly, coup values are:
(10.15)
The coup constraint after trade is, therefore, Vr(C, ϕ) > Vr(D), which can be written as:
(10.16)
It is straightforward to check that the coup constraint after trade, (10.16), binds less often than the coup constraint before trade, (10.13). This occurs for two reasons. First, as shown in the previous subsection, after trade, taxes are lower; therefore, democracy is less costly to the elites. Second, with trade, a coup is more costly to the elites because they are price-takers; therefore, destruction of the fraction ϕ of their assets is not shared with workers. Stated differently, in a closed economy, once the assets of the elites are destroyed, wages fall and the returns to capital and land increase. This implies that income of the elites falls less than proportionately. In contrast, the returns to capital and land are given by international prices in the open economy, so the incomes of the elites fall proportionally as a result of the coup. This is shown mathematically by comparing the right-hand side of ( 10. 16), which is θ(1 — ϕ), to the right-hand side of (10.13), which is θ (1 — ϕ)θ> θ(1 — ϕ). As a result, a coup is now more costly to the elites.
Similarly, we examine the circumstances when promises of policy concessions by the citizens are just sufficient to avoid a coup, given that such policies are implemented only with probability p (i. e., our basic static coup game of Chapter 7). To do this, we calculate the threshold values for the cost of a coup before trade, ϕ*, and after trade, ϕ*, using the best offer that the citizens can promise the elites. For this, the values of the promise of no redistribution (i.e., τD = 0) in a democracy, considering that they are upheld only with probability p, are:
(10.17)
which follow from (9.15). The corresponding values after trade are:
The closed economy threshold value ϕ* is defined by setting Vr(D, τD = 0) = Vr(C, ϕ) and, hence, is given by:
(10.18)
which is naturally identical to (9.17) with ξ = 1. The relevant threshold level after trade, ϕ*, is in turn given by Vr(D, τD = 0) = V’(C, ϕ*); hence:
(10.19)
The same argument we used to show that (10.16) binds less often implies that:
This says that, once the economy is open to international trade, coups have to be less costly to be attractive. Therefore, this comparison establishes the following proposition:
Proposition 10.1: Consider the economic model and the political game described above and define ϕ* by (10.18) and ϕ* by (10.19). Then:
• If ϕ < ϕ*, there are coups both before and after trade opening.
• If ϕ ≥ ϕ*, there are no coups either before or after trade opening.
• If ϕ* ≤ ϕ < ϕ*, there are coups before trade opening but not after.
This proposition, therefore, shows how globalization might help to consolidate democracy. As the discussion suggests, there are two reasons for this. First, under the hypothesis that condition (10.8) holds (the country in question is labor-abundant relative to the world), trade opening increases the returns to labor and therefore to the poorer segments of society, relative to the returns to capital and land. Via this channel, increased international trade reduces equilibrium taxes. With lower taxation, democracy is more likely to survive. We think of this channel as loosely corresponding to a reduction in class conflict between the elites and the citizens. Such conflict is less in a more globalized economy, at least under the assumptions of this standard model.
The second reason relates to changes in the costs of a coup as a result of trade opening. In a closed economy, the costs of a coup are shared between the elites and the citizens because of general equilibrium price effects. More explicitly, the destruction of part of the stocks of capital and land reduces wages and increases the returns to capital and land because capital and land now become “scarcer.” This general equilibrium price effect partly offsets the reduction in the income accruing to capital and land. In an open economy, factor prices are given, and capital owners and landowners bear the full burden of the destruction of their asset stocks, which also tends to make coups less attractive.
4. Political Conflict - Transition to Democracy
That increased international trade makes democracy less redistributive also has implications for the transition to democracy. Recall that a barrier to the transition to democracy is the fear of the elites that democracy will be highly anti-elite. This fear may make them choose repression rather than democratization. If international trade makes democracy less redistributive, it should alleviate concerns of the elites and they may now prefer to concede democracy rather than use repression to quell a potential revolutionary threat.
To analyze the issues, we return to the model of democratization in the presence of a revolutionary threat - specifically, the version used in Chapter 9 in which capital, land, and labor were introduced as three productive factors. The underlying economic model is the same as the one described earlier, and we look at it before and after trade opening. The extensive-form game depicted in Figure 6.2 captures the strategic setup.
We start with nondemocracy and assume that the citizens have a potential revolutionary threat. After revolution and before trade opening, the payoff to the citizens from revolution is:
(10.20)
withgiven by (9.4). After trade, we have instead:
withgiven by (10.11). Both before and after trade, the elites get zero after revolution.
If the elites choose to repress to avoid either a revolution or democratization, we assume they lose a fraction K of their capital and land. This assumption about the costs of repression mirrors our assumptions about the costs of coups. The rest of the setup is the same as before. In particular, values to the citizens and to the elites if there is a democratization are given by (9.7) before trade and by (10.14) after trade.
If the elites choose repression before trade, the payoffs are:
(10.21)
which correspond to (9.8) with= 1.
After trade, the values from repr
ession change in a way similar to the values from a coup. In particular, we have:
(10.22)
Finally, the elites could offer redistribution under the existing regime without democratizing and without resorting to repression. The best they can do in this case is offer redistribution at the favorite tax rate of the citizens, τP; in this case, the values are identical to those in (9.9). Similarly, after trade, we have:
which take into account that after trade, the most preferred tax rate of the citizens is τP, given by ( 10.12).
To simplify the discussion in this section, we assume that the revolution constraint always binds; that is, θ > µ without trade and:
with trade. Moreover, democracy is always (before and after trade) sufficiently redistributive that it prevents a revolution.
More important we assume that:
(10.23)
These conditions imply that promise of temporary redistribution is not going to be sufficient to prevent revolution. This ensures that we are in the part of the parameter space in which the trade-off is between democratization and repression and greater inequality makes democracy less acceptable to the elites (without this assumption, we may be on the other side of the nonmonotonic relationship between inequality and democratization).
With these assumptions, the analysis of the political equilibrium is straightforward. Before trade, the relevant condition for the elites to prefer democracy is:
This condition defines a closed economy cutoff level κ* such that for all κ ≥κ*, the elites prefer democratization to repression. More explicitly, K* is given by Vr(D) = Vr(O |K*), or by:
(10.24)
Similarly, after trade, we need to check that for the open economy:
so that we have a new threshold defined by Vr ( D) = Vr (O |K* ):
(10.25 )
which, of course, is almost identical to the formula in (10.19).
For all K ≥ K*, the elites prefer democratization rather than using repression in an open economy. The same argument as before immediately establishes that:
and for the same reasons. After trade opening, democracy is less costly because the poor now prefer lower taxes, τ p, as given by ( 10.12) rather thanp. In addition, repression is more costly to the elites in an open economy because the costs that stem from the loss in their productive capital and land from a coup are borne only by them. This is, again, because factor prices are given by world prices; therefore, capital and land do not become more valuable after the disruption caused by repression destroys part of them.
This discussion establishes a parallel proposition to Proposition 10.1, as follows:
Proposition 10.2: Consider the economic model and the political game described above and define κ*by (10.24) and K* by (10.25):
• If κ < κ*, the elites use repression to prevent revolution and democratization both before and after trade opening.
• If κ ≥ κ*, there is democratization both before and after trade opening.
• If κ* ≤κ < κ*, the elites use repression to prevent revolution and democratization before trade opening but there is democratization after trade opening.
This proposition shows that for similar reasons to those that allowed globalization to aid democratic consolidation, globalization may also facilitate a transition to democracy. Globalization makes democracy less redistributive and also increases the costs of using force to prevent transitions to democracy. Through both channels, democratization becomes more attractive relative to repression. Consequently, this model suggests that international trade reduces political conflict by reducing inequality and, via this channel, makes democracy more likely.
4.1 Implications of Land Abundance
This analysis is predicated on the assumption that (10.8) holds, which, in practice, implies that the country in question is abundant in labor (and scarce in capital). Although this seems a reasonable assumption for many nondemocratic countries joining the world economy, there are countries such as Argentina and Chile at the beginning of the last century in which the most abundant factor was land. In this case, the exact converse of (10.8) holds, and international trade increases the relative income of the elites.
The implications for the political equilibrium are obvious from this analysis, and we state this simply as the following corollary:
Corollary 10.1: Consider the economic model described above and suppose the converse of (10.8) holds. Then, trade opening makes democratization and democratic consolidation less likely.
This corollary is useful in stressing that the implications of international trade for the political equilibrium depend on its implications for factor prices. Although we emphasized the equalizing role of international trade based on the presumption that labor is the abundant factor in many nondemocratic countries, in certain cases international trade can increase the price of land and the incomes of the elites, thereby potentially making repression and coups more attractive for them. Whether this is so is an empirical question we leave for future research.
5. Financial Integration
Another dimension of globalization is increased financial integration. We now analyze how increased financial integration affects the consolidation of democracy and the likelihood of the use of repression to prevent transition to democracy.
We distinguish between two cases, referred to as the “capital-in” and “capital-out” cases. Capital-in is the usual case in which increased financial integration leads to capital flows toward the capital-scarce country. Capital-out, on the other hand, refers to the case in which capital may fly from the less developed country, despite the fact that the country is more capital-scarce, because of heavy taxation there. We analyze these two cases separately because they emphasize different mechanisms.
5.1 Capital-in and Democracy
Consider the same model as in the previous section, but assume that there is no trade in intermediate goods. Instead, we investigate the implications of factor mobility — specifically, capital mobility. Again, we think that the country in question is less intensive in κ + σ L than the rest of the world, for which the ratio of capital and land to labor is again denoted by ψ.
Now imagine that there is financial integration and this country opens to capital flows from abroad and for now assume that there is no possibility of capital outflows. The only difference from this economic model is that now domestic production of intermediate goods is given by:
(10.26)
where K’ is the amount of capital owned by foreigners invested in the production of capital-intensive goods in this country. We assume that foreign capital can be invested in this country without any costs.
The same arguments now imply that domestic prices are given by:
(10.27)
and factor prices are:
where w’ denotes the wage rate, r’ is the return to capital, and v’ is the rental rate of land all after financial integration. Combining these equations, we obtain:
(10.28)
Because we assumed that the country in question is scarce in capital relative to the world, it is reasonable to expect that K’ > 0, so that with capital account opening, capital flows into rather than out of the country in question. This generally is the case as long as taxes in this country are not too high relative to taxes abroad. To highlight the forces at work in this subsection we assume that foreign capital is excepted from taxation and from the costs of a coup and also ignore taxation of foreign capital abroad. Then the world rate of return on capital is r = θψθ — 1and with K’ = 0, r’ >