«Abstract This paper argues, through investigating a broad set of historical cases and developing a formal theoretical model, that the type of ...»
Security Threats, Enemy-Contingent Policies and Economic
Development in Dictatorships
This paper argues, through investigating a broad set of historical cases and developing a
formal theoretical model, that the type of security threat a dictatorial regime faces has
implications for economic policy making, and consequently economic outcomes. Dictators
who mainly face internal rivals, either contending elites or democratization movements, often
have incentives to conduct policies that are harmful to economic development. However, dictators who mainly face external security threats are more likely to generate economic development-enhancing policies. Type of security threat facing a dictator thus contributes to explaining the large variation in economic development among dictatorships.
Keywords: Dictatorship, Security threat, Economic development, 1
1. Introduction In “Politics”, Aristotle claimed enlightened monarchy was the best government under ideal conditions (Aristotle 2000). But, monarchy easily slides into tyranny. Aristotle thus concluded that more “balanced” forms of government than one-person rule are safer, as they more often provide decent rule under different contexts. Aristotle‟s insight has empirical support when it comes to economic performance: Dictatorships exhibit far more variation in their economic performances than democratic regimes (Rodrik 2000; Besley and Kudamatsu 2007).1 In recent history, there have been authoritarian growth miracles, such as the East Asian Tigers.
However, there have been even more dictatorial growth disasters, like Zaire and Myanmar.
This paper investigates why dictatorships‟ economic policies and performances vary.
Aristotle focused on the personal characteristics of rulers when differentiating between monarchy and tyranny. Jones and Olken (2005) provide solid empirical evidence that dictators‟ personal characteristics matter for economic growth. However, more systemic factors are at least equally important in explaining divergence between different dictatorships.
Below, we show that even if all rulers are self-interested and motivated by staying in office, some will pursue economic development-enhancing policies, while others pursue policies that lead to developmental disasters. Self-interested rulers choose different policies in different contexts; what seems like an “enlightened monarch” may very well be a self-interested dictator. More particularly, this paper argues that an external security threat, i.e. another state, induces self-interested dictators to produce development-enhancing policies. Dictators who only face internal security threats, like democratization movements or contending domestic elites, are less likely to conduct development-enhancing policies. Section 2 presents a literature review and a qualitative sketch of this paper‟s argument. Section 3 discusses several historical cases that illustrate and support the main argument, including a quasi-experimental
Section 4 provides a formalization of the argument. This section also presents additional empirical implications from the model. Section 5 concludes.
2. Literature review and the basic argument Why do some dictatorships turn out to be growth miracles, whereas others turn out to be growth disasters? One explanation is that dictatorship as a category includes several distinct political regime types, ranging from absolute monarchies to one-party states to military regimes (Hadenius and Teorell 2006; see also Linz and Stepan 1996; Wintrobe 1998;
Przeworski et al. 2000). Institutional differences between different dictatorships again matter for economic policies and outcomes. One important question is whether there are institutions in place that constrain the actions of the dictator and his nearest clique, for example a relatively independent judiciary, a rule-following bureaucracy, or even a strong party apparatus. Empirical results indicate that the existence of well-functioning and relatively independent state institutions is very important for dictatorial regimes‟ propensity to generate economic growth (Reference removed). Institutional variation among dictatorships can thus contribute to explaining the large variation in economic outcomes.
However, actors in power can build, reshape or restructure institutions, at least in the long run.
Due to power concentration in dictatorships, institutional structures are not as difficult to change for dictators as they are for democratic governments. The “logic of organizational proliferation” is one of three main dictatorial survival strategies (Haber 2006): Building new organizations that counter the influence of existing ones can enhance the survival probability of an autocrat. Bureaucracies, courts and other institutional structures can be restructured, strengthened or weakened by the political elite to suit their preferences. In the framework developed below, rational dictators engage in different survival strategies, including the
threat-type facing the regime, if we take a medium- to long run-view. For example, the decision to build a “Weberian bureaucracy” and other developmental state institutions may be spurred by the need to deter a foreign enemy. Let us however return to the literature, before we present the main argument.
The political economy literature has generated multiple models where dictators in self-interest promote economic policies that lead to bad macroeconomic results. Generally, dictators may, because of preferences for private consumption or political survival, have incentives to take actions that have negative consequences for their national economies. In Olson's (2003) model, dictators, especially those with short time horizons, expropriate property to maximize personal consumption, thereby reducing the incentives for citizens to work or invest. In Robinson's (2001) and Acemoglu and Robinson's (2006) models, dictators also maximize discounted utility from consumption. In these models, public investment and economic development strengthen opposition groups and reduce leaders‟ survival probability. Leaders therefore reduce the overall size of the economy, among others through cutting public investment, to maximize expected utility from (discounted) private consumption. The core argument is that “while [capital] accumulation may increase total income, it may induce institutional transition which is unfavorable to the autocrat. If a dictator loses political power then he does not gain from development and will oppose it. Thus a dictator may wish to slow accumulation” (Robinson 1998, 24). Dictators have extra strong incentives to underinvest in public goods in natural resource rich economies (Robinson, 2001). Bueno de Mesquita et al.
(2003) assume that political leaders are motivated by surviving in office, and show how dictators with small winning coalitions, especially when combined with a large selectorate, under-provide growth-conducive public goods.2 For these leaders, it is rational to rather
(1990, 1998) models highlight that certain power-motivated dictators invest heavily in repressive capacity. Investment in a repressive apparatus distorts public resources away from more productive projects. In several of the models above therefore, rational dictators concerned with maximizing private consumption or probability of political survival follow policies that are detrimental to overall economic performance. The notion that dictators may reduce internal security threats by promoting bad economic policies is core to the argument of this paper.
Most dictators have historically presided over economically stagnating countries, which contrasts with the decent growth record of most democracies. However, a few dictatorships have performed astonishingly well. What is the explanation? Below, I argue that external security threats may induce dictators to pursue good economic policies. However, there are situations where also dictators facing internal security threats have incentives to promote good policies. The perhaps most obvious example is when dictators‟ survival depend strongly on legitimacy in broader population segments, for example because of a large probability of popular revolution. Economic crises dramatically increase the risk of regime breakdown (Przeworski and Limongi 1997). Overland et al. (2000) constructed a model where regime breakdown in dictatorships are more likely in times of economic crisis. A rational dictator with weak hold on power would therefore generate economic growth in order to survive.
However, there is a difference between avoiding sudden, short-term recessions, which may lead to revolutions (Davies 1962), and generating sustained economic growth over the long term. Growth over many years, of course, alters the level of income dramatically. Empirically, a high level of income is conducive to democratization (Boix and Stokes 2003; Hadenius and Teorell 2005). A dictator may thus reduce the probability of surviving in the long-run, if he
Bueno de Mesquita and Smith‟s (2009) model indicate that due to conflicting mechanisms, the economic policy response to a revolutionary threat may be non-linear. Moreover, the policy response may depend on access to natural resources and aid.
There are other arguments for why some dictatorships generate good economic policy: Besley and Kudamatsu (2007) show that when a winning coalition is relatively autonomous from the dictator, it may chose to pressure the dictator into promoting growth-enhancing policies. Oneparty regimes with relatively strong party apparatuses and weak leaders, such as present-day China, are examples of such regimes. As Bueno de Mesquita et al. (2003) point out, dictators with relatively large winning coalitions, and especially if they have relatively small selectorates, may also follow growth-conducive strategies. However, the size and autonomy of winning coalitions may to a certain extent be endogenous. The decision to broaden the winning coalition through including new groups and the decision to build autonomous strong organizations, for example a bureaucracy dealing with industrial planning or an efficient army organization, may be spurred by the need to deter a foreign enemy. As Doner et al. (2005) show, the East Asian Tigers‟ geopolitical context with severe external security threats was an important reason why these regimes extended their winnings coalition by offering sidepayments to broad population segments, and was also a reason why they institutionalized their state apparatuses in the way and to the extent they did.
Let us now turn more directly to this paper‟s argument: A dictator may face not only internal, but also external security threats. Furthermore, dictators are often strongly motivated by survival in office. As Wintrobe (1998) correctly points out, dictators have heterogeneous motivations. However, staying in office is generally a major concern, partially because many other potential objectives like money, fortune and fame, the promotion of specific interest
political survival when choosing policies. Economic policies can therefore be analyzed as political survival strategies. More specifically, the optimal survival strategy when facing a foreign (external) security threat may likely include strengthening the industrial, administrative and ultimately military capacity of the country. Dictators who face external security threats want to build up military capacity to fight off or preferably deter attacks from foreign adversaries, since a foreign invasion often leads to the demise of the ruling dictator (Bueno de Mesquita et al., 2003). In order to have a strong fighting force, the dictator needs to develop the national economy through industrialization, promotion of technological innovation and diffusion, development of a well-functioning bureaucracy and investments in public infrastructure. Thus, dictators who mainly face external security threats are likely to pursue developmentalist economic policies.
Dictators facing mainly internal security threats are not equally eager at pursuing good economic policies. The literature reviewed above indicated that promoting bad economic policies may be good politics for the dictator, as it may reduce the severity of internal security threats. If the dictator‟s largest threat to staying in power is an aspiring democracy movement, he may rationally choose policies that negatively affect development. For example, as the highly educated tend to be among those with the strongest democratic values (Lipset 1959;
Inglehart and Welzel 2006), a rational, survival-oriented dictator would harshly regulate and limit the content of and access to education, particularly higher education. The argument can be generalized to other internal threats than democracy-movements as well. Expropriating the property of potential opponents, pursuing clientilistic practices and investing heavily in a repressive domestic security apparatus may mitigate internal threats, but all of these strategies have negative economic effects. Conversely, granting free speech and freedom of media may
on civil liberties. However, restricting civil liberties reduces the diffusion of new ideas and technologies into and within the economy. Dictatorships thus have slower technological change-induced economic growth (reference removed). One interesting example relates to communication technologies like the internet and cell phones, which may generate vast economic opportunities. However, as these tools also increase internal security threats, they may be heavily regulated in dictatorships (see e.g. Hachigian 2002). A dictator motivated by political survival, and facing mainly internal threats, will thus have strong incentives to conduct policies that hurt economic growth. The argument above will be clarified in Section 4 through a formal model, but let us first consider some illuminating historical examples.
3. Empirical illustrations of the security threat-type argument