The effects of political and legal constraints on expropriation in natural resource and manufacturing sectors

Abstract

Using panel data for up to 86 developing countries from 1960–2006, this paper investigates the effects of political and legal constraints on the tendency of host governments to expropriate in the natural resource and manufacturing sectors. Consistent with existing research, we find that host governments are more inclined to expropriate in the natural resource sector than they are in the manufacturing sector and that leftist governments are more apt to expropriate than rightist governments. However, we also find that political and legal constraints on host governments lessen the risk of expropriation, but do so with greater force as to the natural resource sector than for the manufacturing sector, and with larger effect as to leftist governments than rightist governments. The results suggest that leftist governments are especially inclined to expropriate in the resource sector unless constrained by political and legal factors from doing so and, because of lower initial capital requirements and increased mobility of assets, all governments, left and right, have less reason to expropriate in the manufacturing sector and therefore are more likely to resist expropriating even in the absence of political and legal constraints.

Keywords

Expropriation foreign direct investment political and legal constraints economic sectors

Introduction

Since the mid-1990s, host government expropriation of multinational corporations (MNCs) has reemerged as a matter of importance and frequency (Hajzler, 2012). Although expropriation is not nearly as common as in the 1960s and 1970s when developing countries appropriated thousands of foreign firms,¹ the recent wave of nationalization has brought renewed interest to understanding the factors that affect expropriation. Some empirical studies have noted the benefits of political and legal checks on host governments for lessening the threat of expropriation on foreign direct investment (FDI). Indeed, several works have shown that veto players promote stability and procedural predictability to government policies that reduce opportunities for expropriation of foreign assets (Graham et al., 2015; Jensen et al., 2012, 2014; Li, 2009).² Political constraints also afford leverage for individuals and domestic groups who profit from foreign investment to pressure lawmakers to refrain from expropriation policies. Additionally, legal constraints containing strong and independent courts can compel host governments to uphold private property rights, a critical factor weighing against the possibility of appropriation (Henisz, 2000, 2004; North, 1990; North and Weingast, 1989).

In contrast to political and legal constraints, the sectoral empirical literature has investigated other factors to explain the determinants of expropriation. Restrepo et al. (2015), for example, maintain that host governments are more likely to nationalize firms that belong to essential sectors of the economy, and confiscate firms in non-essential ones. Hajzler (2012) finds that businesses in resource sectors are more susceptible to expropriation relative to other industries and the vulnerability increases especially during periods when commodity prices are high. Research by Wellhausen (2015) contains sectoral elements and appropriation concerns but the main focus of her study is on firm nationality and contract sanctity.³ Although these studies provide valuable insights on nationalization, none directly address how political institutions impact expropriation for different sectors. In this paper, we study the effects of political and legal constraints on the risk of nationalization for the natural resource and manufacturing sectors. We also investigate the impact of political ideology on expropriation under different sectors, as historically and recently administrations from the left have been more likely to appropriate private assets. Lastly, we study how leftist governments affect sectoral nationalization under different political and legal checks.⁴

We contend that political and legal constraints have differing impacts on the likelihood of expropriation. Building on earlier research,⁵ we argue that the resource sector appears at most risk for host government expropriation because it typically involves investments that require high capital infusions and utilize assets that are relatively immobile, increasing the need for political and legal constraints on government. By contrast, manufacturing investments generally entail lower initial capital requirements and hold fewer ties to a specific location. Both of these factors put them at lower risk of host government nationalization, making checks on the branches of government less important for reducing risk for manufacturing than for natural resources. We also argue that political ideology of the government influences the likelihood for expropriation. Leftist governments are more likely to nationalize foreign firms but especially in the resource sector, for sovereignty and electoral reasons. However, we find that political and legal checks impose limits on leftist governments and their inclination to expropriate resource firms. Using panel data for as many as 86 developing countries from 1960–2006, we find that host governments with greater political constraints, including through multiple veto players and politically independent courts, reduce expropriation for the resource sector but have lesser effect on the manufacturing sector. We also find that leftist governments are more likely to nationalize than rightist governments, and especially for firms in the resource sector, but greater political and legal checks substantially reduce the likelihood that leftist governments will nationalize resource companies.

Our research contributes to the political economy literature in a few ways. First, our research complements previous empirical studies showing that political factors matter for expropriation risk, and adds nuance to the discussion. While earlier work has noted the benefits of political constraints for reducing nationalization risk overall (Graham et al., 2015; Jensen et al., 2012, 2014; Li, 2009), we augment prior studies by showing how the risk associated with the particular sector affects the need for political constraints. Second, the findings build on previous research indicating that the threat of expropriation increases under leftist governments, and especially for resource firms (Jensen et al., 2012; Kobrin, 1980). Third, and most importantly, we show that political and legal checks matter for leftist governments in reducing resource nationalization. The take away from our paper is that host governments in the developing world that have in place institutions that impose political and legal checks on government are more critical for lowering the risk of expropriation in resource sectors as compared to manufacturing, and this is especially the case for leftist governments.

The paper is organized as follows. In section one, we briefly consider the recent nationalization of foreign assets in developing countries. We discuss the importance of political and legal constraints on host governments especially for the resource sector, and particularly for leftist governments, and develop hypotheses in the second section. In the third section, we address issues of model specification. We present our results in section four. Section five concludes the paper.

Expropriation and FDI

Nationalization of FDI has a long history dating back more than a hundred years (Lipson, 1985). The peak of foreign firm expropriation took place in the 1960s and 1970s, when 369 nationalizations occurred in the 1960s, growing to as high as 1,300 in the 1970s (Hajzler, 2012). By contrast, the 1980s and early 1990s witnessed a deep fall in nationalization cases with at most seventeen appropriations arising during the entire decade of the 1980s and only seven between 1990 and 1994. However, the 1995–2006 period observed renewed interest in expropriation, with host governments appropriating eighteen firms between 1995 and 1999 and 92 from 2000–2006 (Hajzler, 2012). The regions most involved in expropriation also have varied somewhat since the 1960s. Latin America and Africa continue to be most prone to nationalize firms, while the Middle East and Asia have expropriated far less often recently than in the past. Host countries in the Middle Eastern countries nationalized 323 foreign firms in the 1960s and 1970s yet expropriated just one firm between 1990 and 2006. Similarly, Asian countries expropriated 413 entities in the 1960s and 1970s, but only appropriated seventeen firms between 1990 and 2006.

The fewer nationalizations in Asia and the Middle East suggest that outright expropriation is no longer the most commonly used strategy and host governments now more frequently employ “creeping expropriation”⁶ strategies, such as restricting currency transfers, limiting asset repatriation, or changing taxes or regulations that benefit host countries at the expense of the profitability of foreign investment. However, the risk still weighs on the minds of foreign investors because of the high costs associated with government takeovers (Eaton and Gersovitz, 1984). Indeed, recent surveys of Chief Executive Officers of US MNCs operating in Latin America indicate that investors focus on risk in their decisions and give prominence to respect for the rule of law and protection of property rights (Biglaiser and Staats, 2010; Jensen et al., 2012).

Hidden within the overall trends on expropriation is that not all economic sectors have been affected equally. As shown in Figure 1, the percentage of total expropriations that occurred in the primary and manufacturing sectors each decade from 1960–2006 varied significantly, with the primary sector most vulnerable to nationalization relative to manufacturing firms.⁷ However, as illustrated in Figure 1, during the nearly fifty-year period, developing countries expropriated manufacturing firms on a regular basis and it was not a rare event.

Figure 1.

Percentage of total expropriations by economic sector (1960–2006).

An interesting phenomenon is that nationalization declined in popularity in the 1980s and early 1990s when privatization of state owned enterprises took off. The small increase of privatization began in the late 1980s–mid-1990s, as developing countries privatized more than 5,000 state-owned firms between 1988 and 1996 (World Bank, 2008). Thereafter, privatization went into slight decline in the 2000s and nationalization of firms began to increase. Although the similarity of time frame between denationalization and renationalization might suggest a connection between the two, this appears unlikely because the state has retaken relatively few private firms, leaving thousands of denationalized firms in private hands. Additionally, developing countries continued to privatize thousands of firms into the 2000s. Thus, the pace of privatization of state-owned firms is not a reliable factor for explaining or predicting the return of expropriation.

Determinants of sectoral expropriation

The prior section provided background about nationalization that offers an opportunity to theorize about expropriation risk as related to specific economic sectors. Building on existing FDI literature, we propose that distinctive characteristics of investments in natural resources make them more vulnerable to expropriation than investments in manufacturing and that the chances of nationalization, most especially in the resource sector, are lower for countries with high political and legal constraints imposed on government.⁸ We also contend that political and legal checks are most important for leftist governments in constraining their leaders from appropriation and particularly in the resource sector.

Because of the risk of adverse action by host countries, political economy scholars have given much consideration to political constraints. Many studies have found that high levels of political constraints in host countries tend to lessen risks to foreign investors.⁹ Other scholars have focused directly on political constraints and expropriation risk, noting that countries with higher political constraints not only reduce the probability of expropriation but also provide opportunities for domestic firms and individuals to lobby host government policy-makers on behalf of foreign investors (Graham et al., 2015; Li, 2009). The presence of MNCs also will often spur strong criticism from domestic businesses, import-competing sectors, and nationalist groups. The diversity of preferences exhibited by domestic interests regarding expropriation of foreign firms combined with added political constraints on the executive branch limit the prospects of expropriation.

Constraints imposed by a strong and independent judiciary also reduce expropriations. Robust and independent courts, so the story goes, tend to constrain abrupt and arbitrary government policies, slowing down what government might otherwise do in the absence of such constraining influence (Landes and Posner, 1975: 882; North and Weingast, 1989: 819). Strong and independent courts also help developing countries build up a reputation of adhering to the rule of law, an important ingredient for promoting investor confidence (Staats and Biglaiser, 2012).¹⁰ Thus, a host country that has a good international reputation because of the strength of its judicial sectors and its long-standing adherence to the rule of law will be less likely to risk damaging that reputation by nationalizing foreign assets.

Although political and legal constraints lessen the chance of appropriation, we argue that the benefits of such constraints vary depending on economic sector. We recognize that firms in all sectors are willing to bear some levels of risk, as greater risk frequently translates into higher investment returns. The issue at hand, however, is not the risk that investors are willing to bear but what actions host governments are likely to take that will affect foreign investments. Absent political constraints, we maintain that host governments are more prone to nationalize resource companies than businesses in the manufacturing sector, and this is a result of the characteristics associated with resource sectors: (1) high initial capital infusions; and (2) immobility of assets. Because of these characteristics, political and legal constraints that bind the hands of the executive are more critical for the resource sector.

Looking at the first characteristic, Vernon (1971) posited that resource investors lose their bargaining advantage with host countries after investment reaches the local market because of the high initial capital requirements.¹¹ MNCs in the energy sector, for example, often need to purchase the rights to large parcels of land and bring expensive equipment and machinery into the host country. In contrast, the initial fixed capital for most manufacturing investments is generally not as high as with natural resources (Kobrin, 1987: 613). There are certainly costs in building a factory, transporting machinery, potentially acquiring a small property, and training personnel. And some manufacturing projects do demand large capital infusions, however, the initial costs are usually much lower for manufacturing firms in comparison to extractive industries. Thus, because of the higher upfront costs, constraints are more important for firms in resource sectors as compared to manufacturing.

Besides the higher capital expenses, resource investments also are rather immobile assets relative to manufacturing factories. As Frieden contends in explaining the greater likelihood that host countries appropriate resource firms compared to manufacturing companies, “the income stream created by a copper mine is specific to the place where the copper is located. The mine, and the resource rents associated with it, can be seized by a host country with relative ease. On the other hand, the income stream accruing to a branch plant of a manufacturing multinational corporation typically is specific to its participation in a global enterprise – it relies on managerial, marketing, or technological inputs available only within the firm. While the host government can seize the factory, it cannot appropriate the rents” (Frieden, 1994: 567–568). There are exceptions, particularly where the manufacturing firms enter the country to gain local market access or where the firms are involved in complex goods, which make it more difficult for them to move without facing high costs. But relative to resource firms, manufacturers still possess greater mobility, lessening the value of political constraints to them as compared to resource firms.

In addition to sectoral characteristics, the increased risk associated with resource investment affirms the importance of political and legal constraints. As the resource curse literature notes, resource firms are more frequently targets for expropriation relative to manufacturing companies because the capital-intensive nature of resource extraction produces fewer positive spillovers to domestic firms and limited benefits to the host country, spurring leaders to expropriate them.¹² Similarly, Jensen and Johnston find that countries with “high levels of natural resource wealth decrease the incentives for politicians to uphold contracts with private investors” (Jensen and Johnston, 2011: 680). We see these explanations as complementary to our argument that host governments that have in place institutions that impose political checks and balances on branches of government are critical for lowering the risk of nationalization in the resource sector. In contrast, such political constraints should hold lesser weight in the manufacturing sector.

Based on the prior theoretical presentation, we propose our first hypothesis:

H1: The higher the level of political and legal constraints on government, the lower the difference in expropriation of firms in natural resources as compared to manufacturing.

We also expect political ideology to have differential effects on expropriation. Historically and recently, governments from the left have nationalized firms but especially foreign resource companies. One reason populist leaders on the left prefer nationalizing in the resource sector deals with sovereignty/patrimonial issues.¹³ Compared to most manufacturing operations, politicians often deliver speeches about the patrimony of natural resources to stir up rally-around-the-flag opposition against foreign investment (Kobrin, 1987: 614). The fact that natural resources come from the land in the host country produces a visceral reaction that the minerals and oil should belong to the people and not some foreign entity. Manufacturing provokes less public reaction, in part, because it is unconnected to the soil and foreign firms bring the operation to the host country.

Leftist governments also frequently treat investment from natural resource firms and manufacturing differently because of the benefits provided to their constituents. Private firms in the natural resource sector tend to produce relatively few jobs for the public at large in host countries. Resource companies typically rely on more capital-intensive production methods and bring in managerial staff from their home country. Mining and petroleum operations, for example, involve the importation of heavy equipment that employ a minimal domestic work force and create fewer linkages to local firms as compared to manufacturing. Indeed, critics commonly register complaints about enclave economies formed by natural resources. Indeed, as Kobrin (1980: 86) notes, resource sectors ramp up the risk of adverse actions. Because of the limited job benefits for domestic employees in the natural resource sector, left-leaning executives generally prefer state ownership, where they can engage in feather bedding, producing more jobs than are necessary for a task, to gain popular support from state employees, an important constituency for the left (Biglaiser and Brown, 2003).

Foreign manufacturing firms, on the other hand, usually depend on domestic workers for their operations abroad. The lower cost of labor in host countries serves as one attraction for foreign companies to open up manufacturing plants. Moreover, and building on work by Jensen et al. (2012: 91), workers in host countries support FDI in sectors that complement labor production, an added plus for manufacturing over resource firms. Given the larger benefits to labor from manufacturing relative to natural resource firms, we argue that leftist governments should increase the risk of nationalization for resource firms.

The increased risk of appropriation by the left is not to suggest that nationalist governments from the right will not also engage in expropriation. But the incentives are likely to be far greater for leftist governments ideologically inclined to favor regulations of private industry and oftentimes domestic government ownership of firms; rightist administrations generally prefer less government involvement in the economy and favor private industry, not public. Many leftist constituencies expect and want domestic government intervention in the economy (Luna and Kaltwasser, 2014).¹⁴ A leftist government can stir up nationalistic fervor and claim credit for creating better jobs for locals when it punishes foreigners by expropriating their firms. This is especially so in the natural resource sector.

H2: Leftist governments are more likely to expropriate natural resource firms than are rightist governments.

Building on the previous theories, we also expect political and legal checks are most critical for leftist governments in natural resource sectors. As noted earlier, historically and recently governments from the left have actively appropriated firms primarily in natural resources. If we consider leftist governments from Latin America, we find many examples from the past (Juan Perón of Argentina in the 1940s; Salvador Allende of Chile in the 1970s; Fidel Castro of Cuba in the 1950s and 1960s; and Lázaro Cárdenas of Mexico in the 1930s) and the last decade (Cristina Fernández de Kirchner of Argentina; Evo Morales of Bolivia; Rafael Correa of Ecuador; and Hugo Chávez of Venezuela) where executives nationalized foreign resource firms. We also see instances where there were few political and legal checks to limit their rule.

Looking at the most recent cases, Argentina, Bolivia, Ecuador, and Venezuela all engaged in the expropriation of natural resources. The countries also all witnessed dramatic declines in the constraints on the executive branch prior to the expropriations and, except for Venezuela, which nationalized a cement and corn processing plant, almost none expropriated manufacturing firms during the period.¹⁵ In the case of Venezuela, President Chávez engaged in widespread nationalization of foreign investment, imposing tougher contract terms on MNCs (The Economist, 2007b), and outright seizing of two oil fields (Li, 2009: 2). Prior to carrying out expropriation in the resource sector, Chávez adopted plebiscitarian strategies that enabled him to close one house of Congress, where his supporters held only about a third of the seats, in an effort to transform the country’s political institutions and concentrate power. Such executive actions reduced political checks on Chávez, allowing him to entrench himself in office, and to take carte blanche with the political opposition and economic interests, including foreign oil investors (Weyland, 2013: 18–19, 23).

Many of Chávez’s political allies in South America including Presidents Fernández de Kirchner of Argentina, Morales of Bolivia, and Correa of Ecuador followed much the same script. The presidents in all three countries “used constituent assemblies to augment executive powers, allow for presidential reelection, and weaken institutional checks and balances” (Weyland, 2013: 19). Indeed, the Morales government blocked the opposition from the critical stages of the drafting of the constitution; Correa engineered the irregular removal of more than half the total members of the Constituent Assembly; and Fernández de Kirchner requested antitrust investigations as well as mob violence against opposition media outlets and special tax audits for businesspeople who publicly criticized her (Weyland, 2013: 22–25). Not coincidentally, executives in all three governments, facing fewer political checks, instituted some form of expropriation/contract renegotiation on foreign energy firms. Argentina expropriated a 51% stake of the Spanish-owned oil company YPF in 2012 (The Economist, 2012);¹⁶ Morales attempted to nationalize his country’s natural-gas industry, only to later sign new contracts with the foreign businesses; and Ecuador’s government confiscated Occidental Petroleum in 2006 (The Economist, 2007a).

Nationalization policies helped leftist presidents gain short-term popular support while also generating revenues for their socio-economic policy goals. The policies, however, may have different impacts on drawing in FDI in specific economic sectors. Indeed, since the expropriations, many of these countries have had difficulties attracting FDI to develop oil and gas reserves (ECLAC, 2014). Between 2006 and 2011, foreign investment in natural resources fell more than 288% in Argentina and 192% in Bolivia. While during the same time period FDI in manufacturing increased by 78% in Argentina and declined by just 16% in Bolivia (ECLAC, 2014: 54).

The recent decline in incoming investment in the primary sector of these countries with reputations for nationalization is not to suggest that the countries in the past upheld high institutional quality. Relatively few political and legal checks existed in the countries when foreign firms first made their initial investments. However, over the years the quality of the institutions subsequently declined, which facilitated expropriation by countries who were not new to nationalization. The point is not that overseas firms simply stay out of low institutional quality countries, as foreign firms frequently invest in relatively unstable developing countries with poorer political and legal checks, but that higher institutional quality prevails upon host countries under leftist governments not to expropriate resource firms.

The previous discussion leads to following hypothesis:

H3: As the level of political and legal constraints increase, leftist governments will be no more likely to expropriate in the resource sector than will be rightist governments.

Research design and methods

To test the hypotheses, we construct country-level panel data using sectoral appropriation for Resources and Manufacturing as dependent variables. We include all available data from at most 86 developing countries between 1960 and 2006 to determine the impact of political and legal constraints on sectoral expropriation. The nationalization data are from Hajzler (2012, 2014). Hajzler supplements his data with appropriation data taken from Kobrin (1980, 1984) for 1960–1979, and from Minor (1994) for 1980–1992. The data are relatively complete and compiled from secondary sources and cover nationalizations that have received the most public attention.¹⁷ Our resource sector model bundles petroleum, mining, and agriculture because all require high capital infusions¹⁸ and involve relatively immobile assets. For the manufacturing sector, we combine all manufacturing firms. The data originally coded sectoral expropriations as counts, but we normalize sector expropriation by dividing by the total number of expropriations (Sectoral Expropriations_ijt/Total Expropriations_jt, where the subscripts i, j, and t, respectively, denotes sector i, country j, and time t) for a fair comparison across different sectors and countries. This approach is consistent with the study by Pinto and Pinto (2008) on sectoral FDI.

For our independent variable measuring political constraints, we employ PolconIII (Henisz, 2000). PolconIII estimates the likelihood of policy change by identifying the number of independent branches of government (i.e., executive and legislative chambers). The expectation is that increased number of government veto powers lead to higher political constraints, which decrease the probability of policy change. PolconIII measures on a continuous 0–1 scale the strength of policy veto points in the government, with higher values reflecting more political constraints. For legal constraints, we use PolconV (Henisz, 2000). PolconV utilizes the same variables as PolconIII, identifying the independent government chambers, but it also includes veto powers from the judicial sector and sub-national units in federal systems. Much like PolconIII, judicial and sub-national veto powers increase the number of legal constraints and inhibit policy change. PolconV is on a continuous 0–1 scale, with higher numbers indicating greater political constraints. Political economy scholarship frequently employs PolconIII and PolconV as proxies for political and legal constraints, respectively (Büthe and Milner, 2008; Jensen, 2008; Li, 2009; Staats and Biglaiser, 2012). We estimate the effects of PolconIII and PolconV in separate models because of multicollinearity between the variables. For our measure of Executive Ideology, we consult Beck et al. (2001) and code leftist executives as 1 and all others as 0.

In addition, and following Li (2009), we consider several possible political and economic control variables. Among political variables, we consider Leader Tenure and Political Stability, all of which the literature predicts lowers the probability of expropriation (Jensen, 2003, 2006; Li and Resnick, 2003; Tuman and Emmert, 2004). We measure leader tenure by calculating the years the current executive’s political party controlled the executive branch. The political stability proxy calculates the number of years the present government has operated under democracy (if at all), assigning a value of 1 if the regime is democratic and 0 otherwise. Cheibub et al. (2009) supply measures of leader tenure and political stability.

We also include controls for economic factors such as Gross Domestic Product (GDP) Growth and GDP per Capita and Trade (as percentage of GDP). There is some disagreement about the effects of economic performance on expropriation risk. One argument is that strong economic growth rates and high levels of development run counter to any supposed economic need to appropriate foreign firms. An alternative argument is that host governments are less prone to nationalize during economic crises because they fear the reputation costs and consequences of nationalizing assets (Jensen et al., 2014). We include economic controls because of their potential impact on expropriation one way or the other. Data for GDP growth and GDP per capita (and its squared term) come from the Penn World Tables (2010). We obtain trade data from the World Bank (2012).

We also add both History of Expropriation and number of Non-Expropriation Years. For history of expropriation, we look at the count of past nationalizations in a country until the past year; this allows us to control for countries that have already seized most of the firms that they find attractive. For non-expropriation years, we include a year counter that records the number of years since the host country last had an expropriation; we also include the cubic polynomials to account for temporal dependence (Carter and Signorino, 2010). These data come from Hajzler (2012, 2014). Following the study by Biglaiser et al. (2016), we also control for potential influences of International Monetary Fund (IMF) programs. Last, we employ Decade Dummy variables to take into account differences in expropriation across time periods, most notably the decline from the earliest years to the latest. We use the decade of the 2000s as our base years. We present variable summary statistics in Table 1.

Table 1.

Descriptive statistics.

Variable	n	Mean	Standard deviation	Minimum	Maximum
Resources	2,991	3.955	18.669	0	100
Manufacturing	2,991	1.251	9.794	0	100
PolconV	2,986	.223	0.278	0	0.89
PolconIII	2,991	0.161	0.198	0	0.688
Executive Ideology	2,991	0.288	0.452	0	1
Gross Domestic Product (GDP) Growth	2,991	1.606	7.737	−43.388	88.748
GDP per capita	2,991	8.072	.936	5.733	11.485
Trade	2,991	64.801	37.788	6.320	280.361
Leader Tenure	2,991	8.094	8.343	1	61
Political Stability	2,991	16.519	13.965	1	109
History	2,991	16.352	36.950	0	300
Non-Expropriation Years	2,991	11.323	9.268	0	43
International Monetary Fund	2,991	31.420	26.282	0	65

A Hausman test indicates between-country heterogeneity and we therefore employ fixed-effects regression for our estimations. We also use heteroscedasticity-robust standard errors.¹⁹ There may be potential concerns about selection bias, as stylized observations suggest that resource firms often invest in poorly governed countries as compared to manufacturing firms that tend to locate in more effectively governed ones. Not to deny stylized facts, but in nearly half the developing countries, host countries expropriated firms in both resource and manufacturing sectors since the 1960s.²⁰ Helping to allay concerns about selection bias, we note a common thread of expropriation in both sectors in so many countries. This suggests that resource firms do not solely seek out poorly governed countries and neither do firms in manufacturing completely avoid such countries. Most importantly, there are only trivial differences in mean scores of PolconV and PolconIII between resource and manufacturing sectors, which suggests that resource investors are no more inclined than manufacturing investors to seek out poorly governed countries.

Results

We report the results of fixed effects models estimating the impact of political constraints and legal constraints of the different sectors in Table 2. Models 1 and 2 address the resource sector and Models 3 and 4 assess manufacturing.

Table 2.

Political and legal constraints, executive ideology, and sectoral expropriations (Fixed Effects Regression, 1960–2006).

Dependent variables:	Model 1	Model 2	Model 3	Model 4
Dependent variables:	Resources	Resources	Manufacturing	Manufacturing
PolconV	−5.128**		−1.451
	(2.116)		(1.030)
PolconIII		−6.701***		−2.156
		(2.328)		(1.337)
Executive Ideology	3.631**	3.734**	2.000	2.042
	(1.784)	(1.794)	(1.249)	(1.260)
Gross Domestic Product (GDP) Growth	−0.026	−0.028	0.025	0.024
	(0.070)	(0.070)	(0.033)	(0.032)
GDP per capita	−7.648	−7.650	−0.505	−0.482
	(15.708)	(15.553)	(4.614)	(4.637)
GDP per capita squared	0.496	0.501	0.032	0.033
	(1.012)	(1.002)	(0.290)	(0.291)
Trade	0.027*	0.025	−0.012	−0.012
	(0.016)	(0.016)	(0.008)	(0.008)
Leader Tenure	−0.013	−0.006	−0.011	−0.009
	(0.059)	(0.058)	(0.024)	(0.024)
Political Stability	−0.055	−0.063	0.001	−0.001
	(0.043)	(0.045)	(0.030)	(0.031)
History	−0.105***	−0.103***	−0.010	−0.009
	(0.026)	(0.026)	(0.047)	(0.047)
Non-Expropriation Years	−0.135	−0.145	−0.551**	−0.555**
	(0.326)	(0.328)	(0.247)	(0.249)
International Monetary Fund	−0.089***	−0.088***	0.003	0.004
	(0.030)	(0.030)	(0.015)	(0.016)
1960s	−0.258	−0.267	0.742	0.723
	(3.146)	(3.131)	(1.317)	(1.324)
1970s	8.052***	8.026***	2.758**	2.728**
	(2.944)	(2.904)	(1.089)	(1.079)
1980s	−0.310	−0.235	0.259	0.264
	(1.847)	(1.848)	(0.805)	(0.812)
1990s	0.166	0.174	0.303	0.304
	(1.196)	(1.238)	(0.552)	(0.555)
Constant	33.838	33.457	4.714	4.511
	(60.432)	(59.934)	(18.048)	(18.178)
n	2,986	2,991	2,986	2,991
R²	0.08	0.08	0.03	0.03
Number of Countries	86	86	86	86

Notes: the results control for country-fixed effects and heteroscedasticity. Non-expropriation years squared and cubed are not reported in table. P-values: ***p < 0.01; **p < 0.05; *p < 0.10. The hypotheses tests are based on two-tailed tests.

Looking first at the resource sector in Models 1 and 2, we find that the coefficients for both PolconV and PolconIII are negatively signed, as expected, and statistically significant at p < 0.05 and p < 0.01, respectively. In calculating substantive importance, we find that an increase of one standard deviation in PolconV leads to a 36% reduction in expropriation, whereas the same change in PolconIII leads to reduction of 33%. Models 3 and 4, covering the manufacturing sector, tell a different story. While the coefficients for PolconV and PolconIII are negative, neither reaches statistical significance. The results provide support for Hypothesis 1 that political and legal constraints lower the risk of expropriation for resource firms as compared to manufacturing.

We also find that executive ideology influences expropriation but more so for the resource firms. In Models 1 and 2, leftist governments are significantly more inclined to nationalize resource firms than rightist governments (p < 0.05). However, executive ideology has less impact on expropriation in manufacturing, failing to achieve statistical significance at even the p < 0.10 level. In terms of substantive significance, using Model 1 and 2, we find that leftist rule leads respectively to increases in resource expropriation of 124% and 129%.²¹ The findings support our second hypothesis that leftist governments are more inclined to expropriate natural resource firms as compared to manufacturing.²²

In Table 3, we show the effects of leftist governments on expropriations in natural resources conditional upon political and legal constraints. As suggested in Hypothesis 3, we test whether such a tendency of leftist governments expropriating more natural resource firms remains, regardless of levels of political and legal constraints, or declines as the constraints levels increase. Models 1 and 2 in Table 3 and the graphs in Figure 2 report the results. First, Models 1 and 2 in Table 3 show that the coefficients of leftist governments take a positive value, indicating that they expropriate firms more often in the natural resource sector, but only when political and legal constraints are absent. The negative and statistically significant signs of the interaction variables between leftist governments and PolconV and PolconIII imply that the higher tendency of leftist governments to expropriate in the resource sector as compared to rightist governments diminishes as political and legal constraints get stronger. The graphs in Figure 2 precisely demonstrate this. The vertical axis in each graph represents the greater tendency of leftist governments to expropriate compared to rightist governments. The horizontal axis in each case represents the degree of political and/or legal constraints on government. As we move to the right on the horizontal axis, leftist governments become essentially equal to rightist governments in their tendencies to expropriate. This trend supports Hypothesis 3. The findings here again emphasize the importance of political and legal constraints in explaining governments’ expropriation behavior.

Table 3.

Political and legal constraints, executive ideology, and expropriations in resource sector (Fixed Effects Regression, 1960–2006).

	Model 1	Model 2
Executive Ideology	6.993**	6.397**
	(2.787)	(2.669)
PolconV	−2.489
	(2.078)
PolconIII		−3.587
		(2.556)
Ideology × PolconV (PolconIII)	−9.235**	−10.778**
	(3.841)	(5.238)
Gross Domestic Product (GDP) Growth	−0.022	−0.028
	(0.069)	(0.069)
GDP per capita	−9.643	−9.559
	(15.550)	(15.439)
GDP per capita squared	0.596	0.606
	(1.001)	(0.993)
Trade	0.027*	0.024
	(0.016)	(0.015)
Leader Tenure	−0.021	−0.010
	(0.057)	(0.056)
Political Stability	−0.053	−0.066
	(0.042)	(0.045)
History	−0.108***	−0.106***
	(0.023)	(0.024)
Non-Expropriation Years	−0.100	−0.137
	(0.315)	(0.324)
International Monetary Fund	−0.089***	−0.090***
	(0.030)	(0.030)
1960s	−0.133	−0.227
	(3.034)	(3.081)
1970s	8.408***	8.221***
	(2.866)	(2.874)
1980s	−0.102	−0.092
	(1.824)	(1.864)
1990s	0.272	0.218
	(1.218)	(1.270)
Constant	42.127	41.255
	(59.804)	(59.590)
n	2,986	2,991
R²	0.08	0.08
Number of Countries	86	86

Notes: the results control for country-fixed effects and heteroscedasticity. Non-expropriation years squared and cubed are not reported in table. P-values: *** p < 0.01; ** p < 0.05; * p < 0.10. The hypotheses tests are based on two-tailed tests.

Figure 2.

Executive ideology, political and legal constraints, and expropriations.

With respect to the control variables, we find, as expected, that a host country with many recent continuous years of not expropriating tends to continue this pattern over time, especially in the manufacturing sector.²³ The negative and statistically significant results for history of expropriation, but only for the resource sector, may suggest that countries already have seized many of the resource firms that they find attractive. Interestingly, trade has a positive effect on expropriation for resources and a negative, but insignificant effect for manufacturing. We do not have a ready explanation for this. Since manufacturing sometimes involves vertical integration, using multiple stages of production operating in several countries, this may reduce the incentive for host countries to expropriate given that a good at a single stage holds limited value on global markets as compared to the nationalization of upstream operations such as is possible in the energy sector. We also find that IMF programs reduce the likelihood of expropriations in the resource sector, a finding consistent with Biglaiser et al. (2016). Most other economic and political controls are in the anticipated direction, but in the majority of instances do not achieve statistical significance.

Conclusion

The recent return of expropriation has brought renewed interest in understanding the determinants of nationalization. Our empirical work is a start to understanding how political and legal constraints affect the risk of nationalization for different economic sectors. Foreign resource companies are particularly vulnerable because their investments typically need large initial capital injections and are fairly immobile assets. Expropriation also poses risks to manufacturing firms but generally not at the same level as natural resources because of the characteristics of the investment – normally smaller initial capital infusions and greater asset mobility. Our results suggest political and legal constraints are more imperative for lessening the risk of nationalization for resources over manufacturing. Our findings also indicate that while leftist governments are more likely to expropriate firms in natural resource sectors than rightist governments, political and legal checks appear to lessen the chance that leftist governments will nationalize in the resource sector.

The findings contained here add to the political economy literature in a few ways. First, our results provide further evidence for the importance of political constraints and politically independent and strong courts but especially for sectors most at risk of expropriation. Previous research indicated the benefits of political constraints on executives for restraining efforts to nationalize FDI.²⁴ Our research bolsters such findings and adds to them. Our results also build on prior sectoral research by Frieden (1994), Hajzler (2012), Jensen et al. (2014), Kobrin (1980), and Vernon (1971), showing variations in expropriation risk associated with different economic sectors. We find that political and legal constraints are critical for the resource sector that confronts the highest risk of nationalization. Perhaps, most importantly, our finding that political and legal checks matter most for reducing the likelihood that leftist governments engage in expropriation in the resource sector adds another layer to the discussion of how higher quality political institutions have varying effects on parties from different ideologies and in different sectors of the economy.

More research is needed on checks and balances and political economy issues. Future studies may want to consider the effects of political and legal constraints on sectoral FDI flows. It is plausible that political constraints influence FDI flows of some sectors more than others, particularly in the sectors that presumably carry the highest risk. Similarly, the effects of checks and balances may also affect the level of privatization in different economic sectors, with risk considerations parried by host country political constraints. The take away from our paper is that host countries in the developing world always possess the nationalization weapon but political and legal factors help to impose constraints on government takeovers of firms most vulnerable to expropriation and this is plausible especially for governments from the left.

Footnotes

Appendix

Table A5.

Sensitivity analysis: political and legal constraints, executive ideology and sectoral expropriations (oil price controlled).

	Model 1	Model 2	Model 3	Model 4
Dependent variables:	Resources	Resources	Manufacturing	Manufacturing
PolconV	−5.125**		−1.426
	(2.125)		(1.018)
PolconIII		−6.962***		−2.175
		(2.356)		(1.340)
Executive Ideology	3.665**	3.782**	1.966	2.009
	(1.797)	(1.806)	(1.230)	(1.241)
Gross Domestic product (GDP) Growth	−0.030	−0.032	0.022	0.021
	(0.071)	(0.071)	(0.033)	(0.032)
GDP per capita	−8.398	−8.385	−0.891	−0.882
	(15.735)	(15.563)	(4.728)	(4.739)
GDP per capita squared	0.528	0.533	0.064	0.066
	(1.018)	(1.007)	(0.297)	(0.297)
Trade	0.027*	0.025	−0.007	−0.008
	(0.016)	(0.016)	(0.008)	(0.008)
Leader Tenure	−0.009	−0.002	−0.008	−0.006
	(0.060)	(0.059)	(0.024)	(0.024)
Political Stability	−0.049	−0.057	0.003	0.001
	(0.042)	(0.044)	(0.028)	(0.029)
History	−0.101***	−0.100***	−0.005	−0.005
	(0.024)	(0.024)	(0.045)	(0.044)
Non-Expropriation Years	−0.166	−0.179	−0.572**	−0.576**
	(0.327)	(0.329)	(0.249)	(0.251)
International Monetary Fund	−0.118***	−0.118***	−0.058*	−0.057*
	(0.043)	(0.043)	(0.034)	(0.033)
Oil Price	−0.029	−0.031	−0.064*	−0.064*
	(0.036)	(0.036)	(0.036)	(0.036)
1960s	−2.699	−2.909	−3.839	−3.884
	(3.562)	(3.521)	(3.009)	(2.999)
1970s	6.410*	6.230*	−0.644	−0.690
	(3.304)	(3.245)	(1.875)	(1.865)
1980s	−0.807	−0.790	−0.695	−0.700
	(1.901)	(1.898)	(0.896)	(0.902)
1990s	−0.089	−0.105	−0.190	−0.192
	(1.220)	(1.256)	(0.630)	(0.630)
Constant	40.170	39.946	10.852	10.729
	(60.572)	(59.967)	(18.358)	(18.429)
n	2,966	2,971	2,966	2,971
R²	0.08	0.08	0.04	0.04
Number of Countries	86	86	86	86

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Aitken

Hanson

Harrison

(1997) Spillovers, foreign investment, and export behavior. Journal of International Economics 43(1–2): 103–132.

Beck

Clarke

Groff

et al . (2001) New tools in comparative political economy: The database of political institutions. World Bank Economic Review 15(1): 165–176.

Biglaiser

Brown

(2003) The determinants of privatization in Latin America. Political Research Quarterly 56(1): 73–85.

Biglaiser

Staats

(2010) Do political institutions affect foreign direct investment? A survey of US corporations in Latin America. Political Research Quarterly 63(3): 508–522.

Biglaiser

Lee

Staats

(2016) The effects of the IMF on expropriation of foreign firms. Review of International Organizations 11(1): 1–23.

Büthe

Milner

(2008) The politics of foreign direct investment into developing countries: Increasing FDI through international trade agreements? American Journal of Political Science 52(4): 741–762.

Carter

Signorino

(2010) Back to the future: Modeling time dependence in binary data. Political Analysis 18(3): 271–292.

Cheibub

Gandhi

Vreeland

(2009) Democracy and dictatorship revisited. Public Choice 143(1–2): 67–101.

Eaton

Gersovitz

(1984) A theory of expropriation and deviations from perfect capital mobility. The Economic Journal 94(373): 16–40.

10.

Economic Commission for Latin America and the Caribbean (ECLAC) (2014) Foreign Direct Investment in Latin America and the Caribbean, 2013. Santiago, Chile: United Nations.

11.

Frieden

(1994) International investment and colonial control: A new interpretation. International Organization 48(4): 559–93.

12.

Gehlbach

Malesky

(2010) The contribution of veto players to economic reform. Journal of Politics 72(4): 957–975.

13.

Graham

BAT

Johnston

Kingsley

(2015) Even Constrained Governments Steal: The Domestic Politics of Transfer and Expropriation Risks. University of Oxford, Blavatnik School of Government, Working Paper, June 2015. Available at: https://www.bsg.ox.ac.uk/sites/www.bsg.ox.ac.uk/files/documents/BSG-WP-2015-002.pdf (accessed 7 January 2017).

14.

Hajzler

(2012) Expropriation of foreign direct investment: Sectoral patterns from 1993 to 2006. Review of World Economics/Weltwirtschaftliches Archiv 148(1): 119–149.

15.

Hajzler

(2014) Resource-based FDI and expropriation in developing economies. Journal of International Economics 92(1): 124–146.

16.

Henisz

(2000) The institutional environment for multilateral investment. Journal of Law, Economics and Organization 16(2): 334–364.

17.

Henisz

(2004) Political institutions and policy volatility. Economics and Politics 16(1): 1–27.

18.

Jensen

(2003) Democratic governance and multinational corporations: Political regimes and inflows of foreign direct investment. International Organization 57(3): 587–616.

19.

Jensen

(2006) Nation-States and the Multinational Corporations: A Political Economy of Foreign Direct Investment. Princeton, NJ: Princeton University Press.

20.

Jensen

(2008) Political risk, democratic institutions, and foreign direct investment. The Journal of Politics 70(4): 1040–1052.

21.

Jensen

Johnston

(2011) Political risk, reputation, and the resource curse. Comparative Political Studies 44(6): 662–688.

22.

Jensen

Biglaiser

et al . (2012) Politics and Foreign Direct Investment. Ann Arbor, MI: University of Michigan Press.

23.

Jensen

Johnston

Lee

C-y

et al . (2014) Crisis and Contract Breach: The Domestic and International Determinants of Expropriation. Working Paper, 20 February 2014. Available at: http://www.natemjensen.com/wp-content/uploads/2014/09/Crisis-Expro-Nov11.pdf (accessed 7 January 2017).

24.

Kobrin

(1980) Foreign enterprise and forced divestment in LDCs. International Organization 34(1): 65–88.

25.

Kobrin

(1984) Expropriation as an attempt to control foreign firms in LDCs: Trends from 1960 to 1979. International Studies Quarterly 28(3): 329–348.

26.

Kobrin

(1987) Testing the bargaining hypothesis in the manufacturing sector in developing countries. International Organization 41(4): 609–38.

27.

Landes

Posner

(1975) The independent judiciary in an interest-group perspective. The Journal of Law and Economics 18(3): 875–901.

28.

Murshed

Tanna

(2017) The impact of civil war on foreign direct investment flows to developing countries. The Journal of International Trade & Economic Development 26(4): 488–507.

29.

(2006) Democracy, autocracy, and tax incentives to foreign direct investors: A cross-national analysis. Journal of Politics 68(1): 62–74.

30.

(2009) Democracy, autocracy, and expropriation of foreign direct investment. Comparative Political Studies 42(8): 1098–1127.

31.

Resnick

(2003) Reversal of fortunes: Democratic institutions and foreign direct investment inflows to developing countries. International Organization 57(1): 175–211.

32.

Lipson

(1985) Standing Guard: Protecting Foreign Capital in the Nineteenth and Twentieth Centuries. Berkeley and Los Angeles, CA: University of California Press.

33.

Luna

Kaltwasser

(eds) (2014) The Resilience of the Latin American Right. Baltimore, MD: Johns Hopkins University Press.

34.

Manger

(2009) Investing in Protection: The Politics of Preferential Trade Agreements between North and South. Cambridge, UK: Cambridge University Press.

35.

Markusen

Venables

(1999) Foreign direct investment as a catalyst for industrial development. European Economic Review 43(2): 335–356.

36.

Minor

(1994) The demise of expropriation as an instrument of LDC policy, 1980–1992. Journal of International Business Studies 25(1): 177–188.

37.

Mosley

(2003) Global Capital and National Governments. Cambridge, UK: Cambridge University Press.

38.

North

(1990) Institutions, Institutional Change and Economic Performance. Cambridge, UK: Cambridge University Press.

39.

North

Weingast

(1989) Constitutions and commitment: The evolution of institutional governing public choice in seventeenth-century England. The Journal of Economic History 49(4): 803–32.

40.

Penn World Tables (6.3) (2010) University of Pennsylvania: Center for International Comparisons of Production, Income and Prices. Available at: http://datacentre2.chass.utoronto.ca/pwt/ (accessed 7 January 2017).

41.

Pinto

(2008) The politics of investment: Partisanship and the sectoral allocation of foreign direct investment. Economics and Politics 20(2): 216–254.

42.

Restrepo

Correia

DCR

Peña

et al . (2015) Expropriation risk, investment decisions and economic sectors. Economic Modelling 48(3): 326–342.

43.

Ross

Mahdavi

(2015) Oil and Gas Data, 1932–2014. Harvard Dataverse. Available at: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/ (accessed 7 January 2017).

44.

Simmons

(2000) Money and the law: Why comply with the public international law of money? Yale Journal of International Law 25(2): 323–362.

45.

Staats

Biglaiser

(2012) Foreign direct investment in Latin America: The importance of judicial strength and rule of law. International Studies Quarterly 56(1): 193–202.

46.

The Economist (2007a) Bolivia: Tin Soldiers. 15 February 2007. Available at: http://www.economist.com/world/la/displaystory.cfm?story_id=8706221 (accessed 25 June 2015).

47.

The Economist (2007b) Venezuela: It’s Our Oil. 28 June 2007. Available at: http://www.economist.com/world/la/displaystory.cfm?story_id=9410681 (accessed 25 June 2015).

48.

The Economist (2010) Expropriations in Venezuela: Full Speed Ahead. 29 October 2010. Available at: http://www.economist.com/blogs/americasview/2010/10/expropriations_venezuela (accessed 25 June 2015).

49.

The Economist (2012) Argentina’s State-Owned Firms: So Far, Not So Good. May 12^th. Available at: http://www.economist.com/node/21554569 (accessed 25 June 2015).

50.

Tomz

(2007) Sovereign Debt and International Cooperation: Reputational Reasons for Lending and Repayment. Princeton, NJ: Princeton University Press.

51.

Tuman

Emmert

(2004) The political economy of US foreign direct investment in Latin America: A reappraisal. Latin American Research Review 39(3): 9–28.

52.

Vernon

(1971) Sovereignty at Bay: The Multinational Spread of U.S. Enterprises. Cambridge, MA: Harvard University Press.

53.

Wellhausen

(2015) Investor−state disputes: When can governments break contracts? Journal of Conflict Resolution 59(2): 239–261.

54.

Weyland

(2013) Latin America’s authoritarian drift: The threat from the populist left. Journal of Democracy 24(3): 18–32.

55.

World Bank (2008) Privatization Database. Washington, DC: The World Bank. http://data.worldbank.org/data-catalog/privatization-database (accessed 25 June 2015).

56.

World Bank (2012) World Development Indicators. Washington, DC: The World Bank.

57.

Zheng

(2011) Credibility and flexibility: Political institutions, governance, and foreign direct investment. International Interactions 37(3): 293–319.