Immigration, Voting and Redistribution: Evidence from European Elections

By Simone Moriconi (IESEG School of Management and LEM), Giovanni Peri (University of California Davis), and Riccardo Turati (Université Catholique de Louvain)

The idea that political support for redistribution and public good provision is lower in societies with high exposure to immigration and diversity is widely acknowledged. In the US, the great migration inflows of the early 20th century, despite having positive economic effects, reduced public support for the provision of public goods.[1] Similarly in Europe, the inflow of low skilled immigrants during the 2000s weakened citizens’ preferences for redistribution.[2] Natives generally perceive immigrants as a burden to their country’s welfare system, especially if their income is lower, and if they choose their location based on the generosity of the local welfare state (the so-called ‘welfare magnet theory’).  Recent experimental evidence confirms these findings and demonstrates that immigrants’ characteristics are important in affecting the link between immigration and natives’ attitudes towards the welfare state.[3] For instance, high-skilled immigrants, who are most often perceived as net contributors to the welfare state, do not generate anti-welfare attitudes in natives.

Given this evidence, it is important to see how the effect of immigration on preferences translates into policies. A straightforward way is to look at the effect of immigration on voting outcomes. This is the objective of our recent paper, which discusses how migration affected voters’ political preferences over welfare state and public education expansion over 28 elections in 12 European countries between 2007 and 2016.[4] For this analysis, we have primarily used data from the European Social Survey (ESS). These extremely detailed individual data include the name of the party voted for by survey respondents at the last national election. Thus, we selected countries characterized by at least 2 elections during the sample period and linked to each party a synthetic measure, drawn from the party’s political manifesto, of that party’s support for an expansion of the welfare state or for public education. Finally, we merged these data with the stock of immigrants, classified by their education level, who were resident in the region of the survey respondent in the year of the election.

Using these data, we show that voting outcomes are significantly related to the local exposure to immigration, with different effects of low skilled and high skilled immigrants. An increase of high skilled immigration to the region makes individual voters more prone to vote in favour of parties with a political manifesto favourable to welfare and public education expansions. This remains true after accounting for confounding factors, most importantly omitted variable bias driven by the fact that migrants may choose to go to richer regions, and where natives are more pro-redistribution. Our dataset also allows us to show that the effect of high skilled migration on voting outcomes varies with voters’ characteristics. Figure 1 shows the effect is generally larger among the less educated, male voters, residing in small rural areas and it is concentrated either in the younger cohort of natives (below 38 y.o) or the older ones (above 58 y.o.). Educated, female, prime-age and urban voters don’t seem to respond with the same intensity.

Figure 1 : New Welfare State Expansion and Natives’ Individual Characteristics

Source: Moriconi, Peri, and Turati (2019). The graphs plot the coefficients for the effects of the share of high skilled and low skilled immigrants on the indicator of preference for a net welfare expansion. Coefficients are estimated on different subsamples by individual characteristics of natives: education (a), age (b), gender (c) and location (d). All the coefficients are estimated with IV estimations. The shadowed area represents the 95% interval of confidence. All the regressions include individual and regional controls, NUTS2 and year fixed effects. These are the authors’ calculations on ESS, Manifesto Project Database and Eurostat data.

What about the effects of low skilled migration? As for the welfare state dimension of redistribution, low skilled migration does not seem to have much effect on the voting behaviour of natives (see Figure 1). However, we find that natives respond to low skilled immigration by voting for parties that propose to reduce public education: natives may prefer to reduce spending on local public goods fearing that these will benefit primarily less-skilled (and low-income) immigrants.[5] And public education could be benefitting especially newly arrived migrants, who are younger and have more children.

Are the estimated effects sizeable ones? Our estimated coefficients suggest that an increase in the stock of high skilled immigrants by 3 percentage points implies a shift of voters from a moderate Centrist-agenda on education provision and redistribution to the more progressive type of political platforms that characterize most Socialist parties in Europe. An effect quantitatively similar, but with an opposite sign (i.e. a switch towards parties less-favourable to redistribution) is induced instead by an increase in the share of less-educated immigrants by 6 percentage points.

We also perform a complementary exercise to check whether country-level immigration induces parties themselves to change their political agenda. We find that an inflow of low skilled migrants induces parties in a country to propose a political manifesto less favourable to welfare state expansion. This suggests that not only voters, but also parties are to be considered “political agents”, as they adjust their attitudes towards redistribution. As shown in Figure 2 below, this is not only the case for conservative types of parties, but also for some “pro-redistribution” parties such as the Christian Democrats. Our findings suggest that in response to low skilled immigration, European countries’ party systems may have evolved over time to support smaller government size, and this would amount to a change in policy preferences revealed by parties even if voters did not much change their vote across parties.

Figure 2: Parties Families – Average position and Variation
Source: Moriconi, Peri and Turati (2018). The top panel plots the average position of political families, using the initial and the average political position of parties. Initial position is measured by the manifesto in the last available election before the start of the sample period, or the first year when the party is present in the political arena during the sample period (2007-2016). Average political position is measured as the average political manifesto of a party during the sample period 2007-2016. Number of parties in each family are reported over the bars. The bottom panel plots the variation of political families average position regarding welfare state expansion keeping a balanced sample of parties.


Alesina, A., A. Miano & S. Stantcheva (2018). “Immigration and Redistribution.” NBER Working paper 24733.

Card, D., C. Dustmann & I. Preston (2012). “Immigration, Wages and Compositional Amenities.” Journal of the European Economic Association 10(1): 78–119.

Moriconi S., Peri G. and R. Turati (2019). “Immigration and Voting for Redistribution: Evidence from European Elections.” Labour Economics 61 (2019) 101765.

Murard, E. (2017). “Less Welfare or Fewer Foreigners? Immigrant Inflows and Public Opinion towards Redistribution and Migration Policy.” IZA DP No. 10805.

Tabellini, M. (2020). “Gifts of the Immigrants, Woes of the Natives: Lessons from the Age of Mass Migration.” Review of Economic Studies 87(1): 454–486,


[1] See Tabellini (2020)

[2] See e.g. Murard (2017)

[3] Alesina et al. (2018)

[4] Moriconi et al. (2019)

[5] Card et al. (2012)

Economic Determinants of Multilateral Environmental Agreements

By Tibor Besedeš (Georgia Institute of Technology), Erik P. Johnson (Carthage College), and Xinping Tian (Hunan University)

Multilateral environmental agreements come in many flavors. Between 1950 and 2012 over 1100 such agreements have been negotiated between countries. These agreements cover a variety of issues including newer concerns such as global warming and climate change as well as older ones such as acid rain, degradation of habitats, and overfishing. The large number of agreements points to a telling difference between environmental agreements and another type of agreement countries often negotiate, namely the one dealing with international trade. A pair of countries tends to negotiate a single trade agreement that is either comprehensive and covers the entirety of trade between the two (such as NAFTA) or excludes some goods from coverage (many agreements Japan negotiates do not cover agriculture). In the environmental arena, agreements tend to be more issue-specific resulting in the same pair of countries signing many more agreements. For example, prior to 1990 France had ratified 213 multilateral agreements. Among these 179 are between France and Germany, its EU partner, with the agreement underpinning the EU being the sole agreement covering international trade between the two. France and Mexico are both parties to 69 agreements, while prior to Slovakia’s entry into the EU, France had no environmental agreements with Slovakia despite their close proximity to each other. The reasons for the different number of agreements are varied. Some have to do with proximity. France and Germany are neighboring countries and have entered into agreements to deal with transboundary environmental issues such as management of resources spanning the common border or pollution that straddles the border. Given Mexico and France are separated by the Atlantic Ocean, there are significantly fewer environmental issues the two have in common and hence fewer agreements that both are parties to.

The distance between France and Germany on one hand and France and Mexico on the other plays an important role in determining what kinds of environmental agreements they enter. The common border shared by France and Germany generates transboundary issues that need to be managed, be they resource management or habitat preservation. Mexico and France have no such issues. Rather, Mexico and France are much more likely to be signatories of large multilateral agreements that many or all countries in the world sign on to, such as the Montreal Protocol (negotiated to phase out production of substances that deplete the ozone) or the Kyoto Protocol (negotiated to reduce greenhouse gases). France and Germany are also party to such agreements, but also are parties to more agreements that are small in nature, in terms of the number of signatories. These small agreements are between fewer countries as they are designed to deal with specific issues affecting a small set of countries near each other that share common pool resources that need to be managed.

The number of signatories to an agreement is also an important variable, as it may determine how effective and stable agreements are. Several theoretical papers have shown that self-enforcing environmental agreements between many countries will arise when the difference in net benefits between the non-cooperative and fully cooperative outcomes is very small.[1] In other words, large agreements will be formed only when commitments that countries must agree to are small or non-existent. On the flip side, small agreements, between few signatories can be much more effective at dealing with a variety of environmental issues.[2] In other words, large agreements may be an expression of a desire to do something about an issue at some point but entail no immediate commitment to act. Small agreements are more likely to contain binding commitments. Large agreements may be examples of what Scott Barrett characterizes as consensus agreements, which are “broad but shallow”, rather than “narrow but deep,” which tends to be a more apt description of small agreements.[3]

The main thrust of our recently published paper in this area is to understand the economic determinants of environmental agreements.[4] A common concern attributed to many environmental agreements is that they will usually result in additional regulation and new limits on economic activity resulting in economic losses. As such, it is important to understand the extent to which economic factors play a role in countries agreeing to sign an environmental agreement. Given the above discussion, we hypothesize that economic determinants play an important role in determining two countries becoming parties to a small agreement with few other countries. However, when it comes to large, globe-spanning agreements, we hypothesize that economic factors play no role in determining whether two countries sign it.

We use data on multilateral environmental agreements from Ronal Mitchell’s International Environmental Agreement Database Project (2002-2015) on agreements signed between 1950 and 2012.[5] We use these data to examine two aspects: the likelihood that a pair of countries signs a (new) environmental agreement and the number of agreements the pair is currently a party to. We use three groups of explanatory variables. The first are variables used in the international trade gravity literature: sum of GDPs, difference in GDPs, distance, and common language. The second are economic integration variables: sum of imports and existence of a trade agreement between the two countries. The last group of variables are proxies for common pool resources: the length of border between the two countries, whether they are in the same geographic region, or whether they are in neighboring regions. We separately analyze agreement with less than 26 signatories, the median number of signatories of all agreements in our sample, and agreements with more than 68 signatories, which is the 75th percentile in the distribution of the number of signatories across all agreements.

The first step in our analysis is to show that proxies for common pool resources are good proxies. We do so by collecting data on two types of common pool resources we could find that span across countries, namely aquifers and transboundary waters. In both cases, countries that have a longer border, and/or are in the same or neighboring regions, are more likely to either share aquifers or transboundary waters.

In the second step we examine the determinants of the likelihood that countries enter into an environmental agreement. A pair of countries with larger joint GDP but with smaller differences between them (i.e., countries more similar in economic size), that trade more, that are closer in distance, that have a longer border, that share a common language, that are in the same or neighboring regions, and that have a trade agreement are more likely to sign an environmental agreement that has a small number of signatories in virtually any year of our sample. These same factors also cause a pair of countries to be parties to more agreements, except for common language and being in neighboring regions which do not have significant effects. When it comes to large agreements, these same factors either do not have significant effects, or their effect decreases over time.

The conclusion we reach based on our results is that economic factors play a role in determining whether a pair of countries signs a small agreement. In these cases, economic factors matter as small agreements have bite: they usually come with binding commitments that often manifest themselves in costs. On the other hand, when it comes to agreements with many signatories, economic factors play a minor role at best because such agreements entail no costs making it easier for countries to join. The lack of costs associated with the agreement implies that there are few economic considerations countries have to worry about when joining such agreements.


Asheim, Geir B., Camilla Bretteville Froyn, Jon Hovi, and Frederic C. Menz (2006), “Regional versus Global Cooperation for Climate Control,” Journal of Environmental Economics and Management, 51 93-109.

Barrett, Scott (1994), “Self-Enforcing International Environmental Agreements,” Oxford Economic Papers 46, 878–894.

Barrett, Scott (2002), “Consensus Treaties.Journal of Institutional and Theoretical Economics 158 (4), 529-547.

Besedeš, Tibor, Erik P. Johnson, and Xinping Tian (forthcoming), “Economic Determinants of Multilateral Environmental Agreements.” International Tax and Public Finance, doi: 10.1007/s10797-019-09588-z

Gelves, Alejandro, and Matthew McGinty (2016), “International Environmental Agreements with Consistent Conjectures.” Journal of Environmental Economics and Management 78, 67-84.

Hovi, Jon, Hugh Ward, and Frank Grundig (2015), “Hope or Despair? Formal Models of Climate Cooperation.” Environmental and Resource Economics 62(4), 665-688.


[1] See Barrett (1994) and the references therein.

[2]  See Asheim et al. (2006), Gelves and McGinty (2016), and Hovi et. al. (2015).

[3] See Barrett (2002).

[4] Besedeš, Johnson and Tian (forthcoming).


Office-Selling, Corruption, and Long-Term Development in Peru

By Jenny Guardado (Georgetown University)

The idea that colonial institutions are fundamental in explaining the divergent development trajectories of New World countries is well-established.[1] In this view, colonial institutions led to longstanding differences in the protection of property rights or the provision of public goods in countries such as the US and Canada on the one hand and a number of Latin American countries on the other, affecting economic growth.[2] Yet, some of the most damaging economic consequences of colonialism also came from individuals’ intent on extraction for self-enrichment purposes, even within the same institutional framework.  For instance, in Spanish America, the figure of corregidor or provincial governor, is associated with some of the worst abuses committed against the indigenous population.[3] For this reason, it is important to look into the selection and quality of colonial officials as a key mechanism to understand long-run economic divergence, while holding the type of colonial institutions constant.

In a recent paper, I do precisely that.[4] Relying on a unique market for colonial offices in the seventeenth and eighteenth century Spanish Empire, I show that at least part of the impact of colonialism can be explained by their role in attracting certain types of individuals to serve in the colonial government. To do so, I collected an original dataset of the prices paid for provincial governorships (corregidores) between 1670 and 1750 to distinguish individuals seeking office for extractive purposes and investigate their long-run impact on economic development within Peru. As a market-based measure of profitability, office prices offer a unique opportunity to distinguish empirically which positions had higher returns to extraction. Furthermore, to avoid concerns of the Crown selectively timing sales when prices are high, I always compare prices at times in which Spain is involved in European Wars – thus less selective and more fiscally pressured – across provinces that only vary in the potential for extraction.  Figure 1 below shows the geographical distribution of the provinces’ prices paid between 1670 and 1750, mapped onto current boundaries in Peru.

Figure 1. Current Districts and Office Prices

Results show that office prices were much higher at times in which the Crown relaxed its selection criteria —during fiscal crises caused by European wars—in provinces with greater potential for profit vis-a-vis others. On average, prices were 16% higher in provinces with greater potential to profit from a key extractive activity (known as repartimiento or the forced sales of goods at markup prices) relative to others. This result translates into more than three times the yearly wage of a military captain in the Spanish army at the time. Alternative explanations such as prices reflecting changes in the attractiveness of Peruvian provinces during European wars, or selectivity in sales by the Crown, among others, are not borne out in the data.

Rather, additional analyses using individual buyers’ traits show that provinces with greater opportunities for extraction were more likely to be purchased by “worse” individuals – particularly when the Crown was less selective. In other words, individuals of lower social status, less bound by social and reputation costs, were more likely to pay higher prices to purchase positions with greater returns to extraction. Because social capital and reputation were key mechanisms to enforce compliance with the Crown’s interests in 18th century Spain, these individuals were of plausibly lower quality than those whose career (e.g. military) or social capital (e.g. nobility) were easier to monitor and screen by the Crown. In this sense, office selling allowed a relatively “worse” class of official to rule in the Spanish Empire. Although the practice ended formally in 1750, by then the Crown had sold so many appointments in advance, that buyers were still ruling Peruvian provinces well into the 1770s – on the eve of the nineteenth century independence movements.

The next question is: did the rule by these individuals influence the long-run economic prospects of these provinces? The answer is yes.  Estimates show that a larger gap or difference in office prices at times of low oversight (during wars) relative to periods of high oversight (during peace) is associated with lower household consumption, years of schooling and public good provision. Because province fundamentals – that may influence long-run development – do not vary between war and peace times in Europe, price differences likely capture shifts in the selectivity criteria of the Crown due to fiscal considerations and not other factors. Figure 2 below shows the relationship between the gap in office prices and household consumption.  Importantly, this economic gap is already present in 1827 — just after Peru gained independence—suggesting the importance of colonial rather than postcolonial factors.

Figure 2. Office Prices Differences and HH Consumption

Each scatter dot represents the mean of the outcome of interest within each bin plotted against the mean value of the difference in office prices within each bin. The solid line shows the best linear fit using OLS.

One important reason why we observe these effects today is because extraction during this period led to spontaneous rebellions which were usually brutally put down. Detailed data on local rebellions for eighteenth century Peru show that provinces with higher prices paid during European wars relative to peace times experience a higher number of spontaneous uprisings against their colonial rulers in the office-selling period (1673–1751) than immediately afterwards (1752–1780). Furthermore, this relationship is still visible in recent times: districts with higher prices in the seventeenth and eighteenth centuries also exhibit greater initial support for anti-government Maoist guerrillas (Shining Path) in the 1980s.

Similarly, political violence also led the indigenous population to limit intentionally its interactions with the Spanish and mestizo world. Data from two centuries show that in provinces with higher provinces the ethnic segregation of the indigenous population started to become visible post office-selling (1780), worsened in the nineteenth century (1876), and is even higher in contemporary times (2013). While limiting interactions may have served to “protect” the community, it might have also reduced the gains from participating in the market.

Put together, these results show the importance of appointment mechanisms and the quality of colonial officers to understand the impact of colonialism on economic development, even for cases sharing the same institutional framework.


Acemoglu, D., Johnson, S., & Robinson, J. A. (2001); “The Colonial Origins of Comparative Development: An Empirical Investigation. American economic review91(5), 1369-1401.

Banerjee, A., & Iyer, L. (2005); “History, Institutions, and Economic Performance: The Legacy of Colonial Land Tenure Systems in India.” American economic review95(4), 1190-1213.

Dell, M. (2010); “The Persistent Effects of Peru’s Mining Mita.” Econometrica78(6), 1863-1903.

Engerman, S. L., & Sokoloff, K. L. (1997); “Factor Endowments, Institutions, and Differential Paths of Growth Among New World Economies.” How Latin America Fell Behind, 260-304.

Guardado, J. (2018); “Office-Selling, Corruption, and Long-Term Development in Peru.” American Political Science Review, 112(4): 971–995.

Juan, J. (1826). Noticias Secretas de América, Sobre el Estado Naval, Militar, y Politico de Los Reynos del Perú y Provincias de Quito, Costas de Nueva Granada y Chile. (Vol. 2). Taylor.

Moreno Cebrián, A. (1977); El Corregidor de Indios y La Economía Peruana del Siglo XVIII: Los Pepartos Forzosos de Mercancias. Editorial CSIC-CSIC Press.


[1] Engerman and Sokoloff (1997) and Acemoglu, Johnson and Robinson (2001).

[2] Banerjee and Iyer (2005) and Dell (2010).

[3] Juan (1826) and Moreno Cebrián (1977).

[4] Guardado (2018).

Price, Product Quality, and Exporter Dynamics: Evidence from China

By Joel Rodrigue (Vanderbilt University) and Yong Tan (Nanjing University of Finance and Economics)

For the typical Chinese exporter, foreign sales grew exponentially after China’s entry to the WTO.  How was this so-called ‘economic miracle’ achieved in such a short span of time? Answering this question has been the focus of policymakers, government officials, and academic researchers across the globe.

Our recent paper adds to a rich literature studying the determinants of Chinese export growth.[1] In particular, we examine the impact that consumer loyalty has on the market strategies adopted by Chinese firms to successfully grow into high-value export markets.  Even if they were aided by falling trade costs, convincing foreign consumers to purchase Chinese goods for the first time is no small feat.  To clear this hurdle we argue that Chinese firms systematically chose to enter markets producing low quality products and setting low prices.

This doesn’t mean the stereotype that Chinese exports are broadly low price or low quality is accurate.  Rather, as foreign consumers adopted new Chinese goods, producers adjusted production to produce higher quality, higher price, higher-value varieties.  In this sense, the rapid Chinese export-driven economic growth has occurred alongside an observable rise in the nation’s firm-level climb up the value-chain.

Our approach builds on the static O-ring models of endogenous quality choice under monopolistic competition.[2]  We extend this setting to consider the dynamic pricing and product quality decisions by bridging this framework with models of habit persistence and demand accumulation.[3]  A key outcome from this marriage of ideas is that exporting firms will alter markups and product quality over time in order to grow sales rapidly during the initial years after entry and develop a large customer base.

The degree to which firms care about the future, however, depends on the firm’s long-run outlook in competitive export markets.  Small, unproductive firms are more likely to produce low quality products and yield little discount initially because they don’t expect to serve the same consumers more than once.  In contrast, the most efficient firms optimally aim to reach consumers by offering good value for their dollar: high quality products at a relatively low price.

To fix ideas we focus on the production of electric kettles, a small electronic appliance, typical of much of China’s export growth.  In Figure 1 we depict the evolution of export prices, product quality and export sales for a representative firm in a typical export market.  New Chinese exporters, despite producing low quality varieties, sell these goods one percent cheaper than established firms selling the same quality of electric kettle.  While this difference might sound small, it is worth at least 4 percent higher export sales in the firm’s first year – often the difference between breaking even or losing money when a firm enters new markets.

Over time the impacts are even larger.  While the export sales of a typical kettle producer grew by nearly 80 percent between 2001 and 2006, Chinese producers systematically added new features to their products: stainless steel casings, rapid heating systems, larger capacity, etc.  Adding these desirable product characteristics, however, is not free.  Rather, we document that over a five-year period typical input costs rose by twelve percent to incorporate higher quality attributes.

Not surprisingly product quality upgrading is likewise found to drive a large part of Chinese export growth; observed product improvements account for at least 17 percent of the aggregate growth in kettle exports over the same time period. As Chinese firms further entrenched themselves in foreign export markets, their profits rose accordingly: markups increased by nearly 2 percent over the same time period as Chinese firms exported kettles at higher and higher profit margins.

The consequences for trade policy are manifold.  In particular, price effects are likely to be muted in response to changes in trade policy.  While tariff declines are found to directly reduce the price consumers pay, product quality upgrading has an opposing effect.  Quality upgrading offsets price declines because high quality products are more expensive to produce, but also because it induces higher producer markups.

In contrast to the long march towards free and unfettered trade after WWII, recent tariff policy can broadly be characterized by unprecedented tariff increases in many countries.  Nowhere is this more evident than US trade policy vis-à-vis China where tariffs have increased sharply, but – surprisingly – prices have remained remarkably stable despite the rise in trade costs.[4]  Could this reflect changes in product characteristics and markups?  That remains unclear.  What is clear is that Chinese producers will react to tariff change on multiple fronts to maintain and grow their foothold in US export markets.  Ignoring the multi-dimensional responses of producers can potentially misrepresent the nature and impact that tariff policy has on firm behavior.


Amiti, Mary, Stephen J. Redding, and David Weinstein, (2018); “The Impact of the 2018 Trade War on U.S. Prices and Welfare.’’ NBER Working Paper No. 25672.

Fajgelbaum, Pablo, Pinelopi Goldberg, Patrick Kennedy, and Amit Khandelwal, (2019); “The Return to Protectionism.” NBER Working Paper No. 25638.

Gilchrist, Simon, Raphael Schoenle, Jae. W. Sim, and Egon Zakrajsek, (2017); “Inflation Dynamics During the Financial Crisis.” American Economic Review, 107(3): 785-823.

Kugler, Maurice and Eric Verhoogen, (2012); “Prices, Plant Size, and Product Quality.” Review of Economic Studies, 79(1): 307-339.

Piveteau, Paul, (2018); “An Empirical Dynamic Model of Trade with Consumer Accumulation.” Working Paper, Columbia University.

Rodrigue, Joel and Yong Tan, (2019); “Price, Product Quality, and Exporter Dynamics: Evidence from China.” International Economic Review, forthcoming.


[1] Rodrigue and Tan (2019).

[2] See Kugler and Verhoogen (2012).

[3] In our model, habit persistence is based on Gilchrist et al (2017), while demand accumulation is based on Piveteau (2018).

[4] See Amiti et al (2018) and Fajgelbaum et al (2019).

Do Macro Production Functions Differ Across Countries?

Markus Eberhardt (University of Nottingham) and Francis Teal (Centre for the Study of African Economies, University of Oxford)

“We compare this [input] index with our output index and call any discrepancy ‘productivity’… It is a measure of our ignorance, of the unknown, and of the magnitude of the task that is still ahead of us.” Zvi Griliches (1961)

It is an unfortunate misconception that the canonical Solow-Swan growth model necessarily implies that all economies in the world, rich or poor, industrialised or agrarian, possess the same production technology.[1]  As the above quote shows there are prominent critics of this assumption while Solow himself suggested that “whether simple parameterizations do justice to real differences in the way the economic mechanism functions in one place or another” was certainly worth ‘grumbling’ about.[2] Nevertheless, the notion that cross-country empirical analysis should, in case of accounting exercises, adopt or, in case of regression analysis, aim to arrive at a common capital coefficient of around 0.3 for all countries is deeply ingrained in the minds of growth economists.

In a forthcoming paper, we take an alternative approach to revisit the issue of whether technology is common across countries.[3] But before we discuss the approach and findings of that paper, we want to briefly illustrate why the assumption of a common capital coefficient is questionable. If the commonly adopted value for the capital coefficient of 1/3 were of great significance then we would expect total factor productivity (TFP) estimates from standard growth accounting exercises to differ substantially once we chose different values for the capital coefficient, in particular for values close to zero. Our data exercises proceed as follows: we first draw a random capital coefficient between zero and one, which we employ to compute annual TFP growth via standard growth accounting. The accounted country-specific TFP growth is then averaged over time and countries are ranked by TFP growth magnitude. This process is then repeated 1,000 times.

An important feature of this analysis and also our study is the focus on manufacturing. The central importance of this industrial sector for successful development is (still) a widely recognised ‘stylised fact’ in development economics. Yet in contrast to the literature on cross-country growth regressions using aggregate economy or agriculture data there is comparatively little empirical work dedicated to the analysis of the manufacturing sector in a large cross-section of countries. If manufacturing matters for development it seems self-evidently important to learn about the production process and its drivers in this industrial sector.

A visual illustration of the results from the 1,000 growth accounting exercises is provided in Figure 1: the ‘Baobab trees of growth accounting’. We split the 48 countries in the sample into 8 groups, based on the rankings implied by their average TFP growth rate when the capital coefficient is 1/3. On the y-axis of each plot we indicate the capital coefficient applied in the growth accounting exercise, while the x-axis indicates the resulting rank of each country from 1 to 48 in terms of average TFP growth. All of the plots show the same pattern, resembling a Baobab tree, with a narrow ‘trunk’ and a spreading out of the upper ‘crown’. We conclude that provided the capital coefficient is assumed common across countries it is entirely unimportant what value is chosen for this parameter, since the productivity rankings based on TFP growth rates are essentially unchanged for ‘reasonable’ values of the capital coefficient. We only see considerable change if the capital coefficient is greater than 0.6, which we know is an unreasonable parameter value. This finding raises serious doubts over the validity of the common technology assumption maintained in these computations.

Figure 1


In our paper, we compare and contrast regression results from different estimators which make different assumptions about production technology and TFP: whether technology differs or is common across countries, whether TFP levels and growth rates differ or are common across countries. While the theoretical literatures on growth and econometrics provide solid foundations for technology heterogeneity as well as the time-series and cross-section properties of macro panel data we highlight in our paper, these have in practice not been considered in great detail in the empirical growth literature. Since most growth economists are not familiar with our preferred set of empirical estimators, we motivate and discuss them in the paper at great length. These methods enable us to model country-specific production technology and time-invariant unobserved heterogeneity across countries (differential TFP levels), but also time-variant unobserved heterogeneity to capture differential TFP evolution over time in a very flexible manner.

Our analysis establishes that assuming a common technology parameter for all countries yields very serious bias in the capital coefficient, with parameter values ranging from 0.6 to 0.8, far in excess of the macro evidence of 0.3. The heterogeneous parameter estimators yield uniformly lower capital coefficients, much more in line with the aggregate economy factor income share data. Based on formal diagnostic testing, our empirical results favour models with heterogeneous technology which account for a combination of heterogeneous and common TFP and reject the notion of common technology. We also carried out a significant number of formal parameter heterogeneity tests which confirmed this result.

Our general production function framework provides a number of alternative insights into macro TFP estimation. Firstly, it seems prudent to allow for maximum flexibility in the structure of the empirical TFP terms: if TFP represents a ‘measure of our ignorance’ then it makes sense to allow for differential TFP across countries and time. Secondly, it further makes sense to keep an open mind about the commonality of TFP: while early empirical models assumed common TFP growth for all countries, later studies preferred to specify differential TFP evolution across countries. We believe the arguments for commonality (non-rival nature of knowledge, spillovers, global shocks) and idiosyncracy (patents, tacit knowledge, learning-by-doing) call for an empirical specification which does not rule out either by construction. Thirdly, an empirical specification that allows for parameter heterogeneity across countries and for a shift away from the widespread focus on TFP analysis and toward an integrated treatment of the production function in its entirety appears to fit the data best. The Baobab Tree accounting exercises illustrate the very limited insights to be gained from the assumption of a common technology framework.

Our analysis represents a step toward making cross-country empirics relevant to individual countries by moving away from empirical results that characterise the average country and toward a deeper understanding of the differences across countries. Further, the key to understanding cross-country differences in income is not exclusively linked to understanding TFP differences, but requires careful consideration of differences in production technology. Since modelling production technology as heterogeneous across countries requires an entirely different set of empirical methods we have focused on developing this aspect in the present paper and have left empirical testing of rival hypotheses about the patterns and sources of technological differences for future research.


Eberhardt, Markus, and Francis Teal (2019); “The Magnitude of the Task Ahead: Macro Implications of Heterogeneous Technology.” Forthcoming in the Review of Income and Wealth.

Griliches, Zvi, (1961); “Comment on An Appraisal of Long-Term Capital Estimates: Some Reference Notes by Daniel Creamer.” Output, Input, and Productivity Measurement (NBER), pp. 446–9.

Solow, Robert M., (1986); “Unemployment: Getting the Questions Right.” Economica, 53(210): S23–34.


[1] We refer to ‘technology heterogeneity’ to indicate differential production function parameters on observable inputs across countries, with unobservables captured as TFP.

[2] See Solow (1986 S23)

[3] See Eberhardt and Teal (2019)

Should the WTO Require Free Trade Agreements to Eliminate Internal Tariffs?

By Kamal Saggi (Vanderbilt University), Woan Foong Wong (University of Oregon), and Halis M. Yildiz (Ryerson University)

At a time when multilateral trade liberalization at the World Trade Organization (WTO) seems to have come to a grinding halt, preferential trade agreements (PTAs) appear to be the only game in town for countries interested in undertaking international trade liberalization. Under the current rules of the WTO, countries entering into a PTA are required to eliminate tariffs on “substantially all trade” with each other and refrain from raising trade restrictions on non-member countries. In the existing literature, Article XXIV has often been invoked as a justification for the assumption that PTA members impose zero tariffs on each other. Though reasonable, this approach masks the incentives underlying the tariff-setting behavior of PTA members and, by design, fails to shed light on the consequences of requiring them to fully liberalize internal trade. In reality, PTA members do not always abide by this restriction.[1] In a recent article, we employ a model of endogenous trade agreements to shed light on the consequences of such non-compliance on the part of PTA members regarding the free internal trade requirement of GATT Article XXIV.[2]

Our conceptual approach follows the recent literature on endogenous formation of trade agreements.[3] In a modified version of the three-country competing exporters framework, in order to draw out the implications of requiring PTA members to eliminate tariffs on one another, we derive and contrast optimal tariffs and equilibrium trade agreements under two scenarios.[4] While members are required to engage in free internal trade in the WTO-consistent scenario, PTA members have the freedom to implement jointly optimal internal tariffs under the unconstrained preferential liberalization scenario. A comparison of these two scenarios delivers several interesting results.

First, we show that if FTA members choose internal tariffs to maximize their joint welfare, they indeed have an incentive to impose positive tariffs on one another. The intuition for this surprising result rests on the interplay between the lack of external tariff coordination between FTA members and the complementarity of imports tariffs. Since FTA members set their external tariffs independently, each member fails to take into account the benefits that its external tariff confers on its partner and thus the individually optimal external tariffs of FTA members are too low from the perspective of maximizing their joint welfare. While coordinating their internal tariffs, FTA members deliberately choose to set positive internal tariffs on each other: doing so commits each of them to a higher external tariff on the non-member country thereby bringing their individually optimal external tariffs closer to jointly optimal ones. To confirm the role that external tariff coordination plays in generating positive internal tariffs within an FTA, we consider a CU formation game where members can coordinate their external as well as internal tariffs. Under such a case, members indeed find it optimal to engage in free internal trade. This result suggests that the free internal trade requirement of Article XXIV is likely to be more binding for FTAs relative to CUs.

The second major insight delivered by our analysis is that requiring PTA members to eliminate internal tariffs benefits the non-member since it leads to lower external tariffs on the part of PTA members. This result is noteworthy since part of the original intent behind the design of Article XXIV may have been to minimize any potential negative effects of PTAs on non-member countries. Ostensibly, this objective was met by prohibiting PTA members from raising their external tariffs on outsiders. However, in our model, PTA members have no incentive to increase their external tariffs on the non-member country anyway. Thus, the Article XXIV stipulation that PTA members cannot raise tariffs on outsiders may actually do little to protect the interests of outsiders. Our analysis demonstrates that, somewhat surprisingly, it is the Article XXIV requirement of free internal trade within a PTA that ends up protecting the non-member as opposed to the restriction imposed on the external tariffs of PTAs.

Our third major result pertaining to the free internal trade requirement of Article XXIV is that having such a requirement makes it harder to achieve global free trade. By lowering the external tariffs of FTA members, the free internal trade requirement of Article XXIV makes it less attractive for the non-member to enter into trade agreements with them. Thus, the free internal trade requirement of Article XXIV might facilitate some degree of free-riding in the WTO system by allowing non-member countries to benefit from reductions in external tariffs of FTA members (that result from their internal trade liberalization) without having to offer any tariff cuts of their own. Thus, our overall message is somewhat nuanced: when circumstances are such that achieving global free trade is not possible, the free internal trade requirement of Article XXIV increases world welfare by lowering tariffs worldwide but, at the same time, it also reduces the likelihood of reaching global free trade.

We also show that our results are robust to two alternative tariff setting scenarios. First, we relax the assumption that countries seeking to form FTAs set their MFN tariffs non-cooperatively. To this end, we allow countries to engage in a limited degree of cooperation and show that our main results regarding the impact of the free internal trade requirement continue to hold even when countries do not set their tariffs in a fully non-cooperative manner. Second, to address the issue of the extent of enforceability of the free internal trade provision of Article XXIV, we examine a scenario where Article XXIV imposes a ceiling on the internal tariffs of an FTA. Under such a scenario, we show that the free-riding incentive continues to be the pivotal force and tighter ceiling lowers the external tariffs of FTA members, making it less attractive for the non-member to enter into trade agreements with FTA members which in turn undermines global free trade.


Bagwell, K. and R.W. Staiger, (1999); “Regionalism and multilateral tariff cooperation.” In John Piggott and Allan Woodland, eds, International Trade Policy and the Pacific Rim, London: MacMillan.

Bagwell, K., C.P. Bown, and R.W. Staiger, (2016); “Is the WTO passé?Journal of Economic Literature 54(4): 1125-1231.

Saggi, K. and H.M. Yildiz, (2010); “Bilateralism, multilateralism, and the quest for global free trade.” Journal of International Economics 81: 26-37.

Saggi, K., W.-F. Wong and H.M. Yildiz, (2019); “Should the WTO require free trade agreements to eliminate internal tariffs? “, Journal of International Economics, 118, 316-30, 2019.

Saggi, K., A. Woodland, and H.M. Yildiz, (2013); “On the relationship between preferential and multilateral trade liberalization: the case of customs unions.” American Economic Journal: Microeconomics 5(1): 63-99.


[1] See Bagwell et. al, 2016.

[2] See Saggi et al. (2019).

[3] See Saggi and Yildiz (2010) for FTAs and Saggi et. al (2013) for CUs.

[4] See Bagwell and Staiger (1999).

Economic Shocks and Crime: Evidence from the Brazilian Trade Liberalization

By Rafael Dix-Carneiro (Duke University), Rodrigo R. Soares (Columbia University), and Gabriel Ulyssea (University of Oxford)

The idea that economic crises can lead to increased crime is far from new, dating back at least to the Great Depression of the 1930s.[1] Such concern is well justified, as crime imposes a substantial welfare cost on society. However, estimating the causal effect of economic conditions on crime and quantifying this relationship has proven to be elusive. Indeed, finding an exogenous variation in economic conditions is quite challenging and there are different potential threats to identification, such as omitted variable bias and reverse causality.[2]

In a recent paper, we overcome these challenges by exploiting the Brazilian trade liberalization of the 1990s, which provides a natural experiment that generated exogenous shocks to local economies in the country.[3] Brazil is a particularly appealing empirical setting, as there is little evidence on the effect of economic conditions on crime in developing countries with a high incidence of crime. In 2013, Brazil was ranked first worldwide in absolute number of homicides (more than 50,000 occurrences per year) and 14th in homicide rates, with 25.2 homicides per 100,000 inhabitants.[4] However, the country is not an outlier within Latin America and the Caribbean: according to the UNODC, 14 of the 20 most violent countries in the world are located in the region. Besides high incidence of crime, these countries also have in common poor labor market conditions, weak educational systems, and high levels of inequality. In such context, adverse economic shocks can have more severe effects on crime, with potentially larger welfare implications.

In our empirical design, we follow the previous literature[5] and exploit two features of the Brazilian context. First, the trade liberalization episode was not only characterized by large tariff reductions – which fell from 30.5% to 12.8% between 1990 and 1994 – but there was also substantial variation in the intensity of tariff cuts across industries. Second, regions in Brazil have very different economic structures and specialize in the production of different baskets of goods. The combination of these two features therefore implies that the trade liberalization leads to very different levels of exposure to foreign competition across regions. For example, Traipu in the state of Alagoas was largely specialized in agriculture, which actually experienced a slight increase in the level of protection (i.e. tariffs). In contrast, Rio de Janeiro specialized in apparel and food processing, both of which experienced substantial tariff reductions. Thus, one could expect Rio de Janeiro to be more adversely affected by the trade opening than Traipu. This reasoning provides the base for our empirical approach, which exploits this exogenous variation in exposure to the trade shock across regions.

We show that regions more exposed to the trade shock – i.e. more specialized in industries facing larger tariff reductions – experienced a relative increase in the number of homicides in the years immediately after the end of the trade liberalization, but the effect completely vanishes in the long run. This can be seen in Figure 1, which shows the differential increase in the logarithm of crime rates in regions facing larger reductions in tariffs relative to regions that experienced lower tariff reductions. This large effect contrasts with those found in the previous literature, which typically shows that worse economic conditions are associated to higher property crime, but find no effects on homicides. However, previous studies have focused on developed countries (Mustard 2010), which have relatively low crime rates and stronger baseline economic conditions (i.e. lower inequality and better functioning labor markets).

Figure 1 Effect of Trade liberalization on Regional Homicide Rates


Having established the overall effect of the trade shock on crime, we use the dynamics of this effect to directly investigate its potential channels. We show that the trade shock substantially affected different potential determinants of crime, such as labor market conditions, public goods provision (public safety and government spending), and income inequality. However, only the effect on labor market conditions (as measured by employment rates) follows the same dynamic pattern as the effect of the trade shocks on crime. Importantly, these two dynamic responses are very different from those observed for other potential determinants, such as public goods provision and inequality. This strongly indicates that the employment rate is the key channel to explain how these local trade shocks affected crime. In the paper, we develop an econometric framework that exploits these different dynamic responses to identify lower and upper bounds for the effect of labor market conditions on crime. We find that employment rates accounted for 75–93% of the observed effect of the trade shocks on crime.

In sum, our results highlight that crime is an important dimension of the adjustment costs to trade shocks. Hence, to the extent that trade opening leads to transitional unemployment, there can be substantial externalities associated to this adjustment process in the form of temporarily higher crime rates. Moreover, our results indicate that employment rates are the key mediating channel of the overall effect of trade opening on crime.

Interestingly, earlier research shows that the long-run employment recovery in Brazil occurred exclusively via informal employment, as formal employment does not recover even 20 years after the trade opening episode.[vi] These results therefore suggest that informal jobs were crucial in keeping individuals away from criminal activities, despite the fact that they might be of lower quality when compared to those in the formal sector. If this is indeed the case, stricter enforcement of labor regulations could exacerbate the response of crime to adverse economic shocks. Put differently, our results suggest that more lax enforcement of labor regulations – and active labor market policies – may help to prevent increases in crime during economic downturns.


Dix-Carneiro, R., R. Soares and G. Ulyssea (2018); “Economic Shocks and Crime: Evidence from the Brazilian Trade Liberalization.” American Economic Journal: Applied Economics10(4), 158-95.

Dix-Carneiro, R., and B. Kovak (2017a); “Trade Liberalization and Regional Dynamics.” American Economic Review, 107(10), 2908-46.

Dix-Carneiro, R., and B. Kovak (2017b); “Margins of Labor Market Adjustment to Trade.” Journal of International Economics, 117, 125-142.

Fishback, P.V., R.S. Johnson, and S. Kantor (2010); “Striking at the Roots of Crime: The Impact of Welfare Spending on Crime During the Great Depression.” Journal of Law and Economics, 53(4): 715-740

King, L (2009); “Statistics Point to Increase in Crime During Recessions [5]”, The Virginia Pilot, 19 January.

Mustard, D B (2010); “How do Labor Markets Affect Crime? New Evidence on an Old Puzzle.” Published in B Benson and P Zimmerman (eds), Handbook on the Economics of Crime, Edward Elgar, Chapter 14: 342–358.

UNODC (2013); “Global Study on Homicide [7]”, United Nations Office on Drugs and Crime.


[1]  See e.g. Fishback, Johnson and Kantor (2010).

[2] Mustard (2010).

[3] Dix-Carneiro, Soares and Ulysssea (2018).

[4] UNODC (2013).

[5] See Dix-Carneiro and Kovak (2017a, b).

[6] Dix-Carneiro and Kovak (2017b).

The Impact of TRIPS and Compulsory Licensing on Developing Country Markets

By Eric Bond (Vanderbilt University) and Kamal Saggi (Vanderbilt University)

The Trade-Related Intellectual Property Rights (TRIPS) agreement of the World Trade Organization (WTO) requires that all WTO members provide a minimum level of patent protection for all types of intellectual property. This requirement has created a problem for developing countries in obtaining access to patented pharmaceuticals, because pharmaceutical companies are reluctant to sell drugs in middle and lower income countries due to the potential negative impact on prices in high income markets. The spillovers can result from the use of reference pricing in high income markets, whereby a high income country government uses an average of prices in other countries to determine the price that a patent holder can charge in its market.  Spillovers can also arise from illegal arbitrage trade.[1]

As a result of these potential spillovers, newly patented drugs may be unavailable or introduced with substantial delays in middle and low income markets.[2] TRIPS does, however, provide countries with the option of issuing a compulsory license (CL) if the market has not been served in a reasonable period of time. A country issuing a CL is required to provide adequate compensation to the patent holder. There have been a number of examples of the use of CLs to obtain access to patented pharmaceuticals by middle and low income countries since the advent of TRIPS, including drugs to treat AIDS, heart disease, and cancer.[3]

How does the requirement of patent protection under TRIPS, combined with the option of issuing a CL if the market isn’t served, affect the welfare of developing countries and patent holders? In a recent article, we address this question using a game-theoretic model to consider a patent holder’s decision of whether it should incur the fixed cost of entering a developing country market.[4] We show how the answer to this question depends on the imitative ability of the developing country to produce copies of the patented product and the level of fixed costs of entry relative to the profits from the market.

Prior to TRIPS, a developing country could obtain copies of patented products from imitators if it did not provide patent protection.  For countries where the cost of entry for the patent holder was high relative to the profits from entry, typically countries with relatively small markets, the patent holders would only enter if patent protection was provided. The country would then have to choose between providing patent protection and obtaining a high cost, high quality product, or not providing patent protection and obtaining a low quality and low cost imitation. The high entry cost countries would only provide patent protection if the quality of imitators was sufficiently low.

In contrast, for countries where the fixed costs were low relative to the profits from entry, the patent holder might still be willing to enter without patent protection if the quality of the imitators was not too high. These countries obtained a double benefit by not providing patent protection: the patented product was obtained at a low price and the copies were also available for those unwilling to pay the price for patented goods.

The absence of patent protection prior to TRIPS made CLs an unnecessary instrument for developing countries, because imitators could produce patented foreign products without requiring a license.  In fact, we show that the option of using a CL could actually make all parties worse off by reducing the incentive of developing countries to offer patent protection. The insight is that developing countries are better off under imitation relative to a CL and therefore have an incentive to preempt the possibility of the patent-holder resorting to a CL by not recognizing the patent. After all, the issuance of a CL is premised on the legal recognition of the underlying patent.

The TRIPS requirement that developing countries provide patent protection made developing countries worse off and patent-holders better off, because it raised prices of patented products by preventing imitators from providing competition for patent holders. The extent to which the option of a CL mitigates the loss to the developing countries from TRIPS depends on the country’s characteristics. For countries with markets sufficiently profitable that the patent-holder would have entered without a patent, TRIPS primarily benefitted patent-holders by eliminating competition from imitators. For countries that would have had to rely on imitators to provide the product prior to TRIPS, TRIPS provides access to the product through a CL. However, the delay required before a CL can be issued means that the country will not obtain access to a copy of the patented product as quickly as it would pre-TRIPS.

Finally, the fact that the patent holder obtains a royalty payment under the CL means that it might prefer a CL to entry if the return from entry is sufficiently low. Thus, the option of a CL could actually cause countries that provided patent protection pre-TRIPS to experience delay in obtaining access to the patented product under TRIPS. It should be noted that since developing countries do not take into account the profits of patent holders in making their decision whether to provide patent protection, the level of protection was below the socially optimal level pre-TRIPS.

We also consider the case in which the government of the developing country negotiates a price ceiling for which the patented product is to be sold in its market. The effect of the CL in this case depends on the relative bargaining power of the two parties during negotiations over the price ceiling. If the patent-holder has all of the bargaining power, then the government is able to use the threat of a CL to lower the price of the patented product. If the country has all of the bargaining power, the royalty payment required by TRIPS benefits the patent-holder by providing a minimum level of compensation that it must receive for entering the market. Thus, the ability to issue a CL primarily benefits the party whose bargaining position during price negotiations is relatively weaker.


Beall, R. and R. Kuhn, (2012); “Trends in Compulsory Licensing of Pharmaceuticals since the Doha Declaration: A Database Analysis.PLos Medicine 9(1): 1-9.

Bond, E. W., and K. Saggi, (2018); “Compulsory Licensing and Patent Protection: A North-South Perspective.Economic Journal 128 (May): 1157-79.

Cockburn, I.M., Lanjouw, J.O., and M. Schankerman, (2016). “Patents and the Global Diffusion of New Drugs.American Economic Review 106(1): 136-164.

Danzon, P., Y. R. Wang, and L. Wang, (2005); “The Impact of Price Regulation on the Launch Delay of New Drugs,” Journal of Health Economics 14: 269-92.

Goldberg, P. K., (2010); “Intellectual Property Rights Protection in Developing Countries: The Case of Pharmaceuticals.Journal of the European Economic Association 8: 326-53.


[1] See Golderg (2010).

[2] See Danzon, Wang, and Wang (2005), and Cockburn, Lanjouw, and Schankerman (2016).

[3] See Beall, R. and Kuhn, R. (2012).

[4] See Bond and Saggi (2018).


Financial Constraints, Institutions and Foreign Ownership

Ron Alquist, (AQR Capital Management), Nicolas Berman (Aix-Marseille University), Rahul Muhkerjee (Graduate Institute, Geneva), and Linda L. Tesar (University of Michigan)

Cross border mergers and acquisitions (CBMA) as a form of foreign direct investment (FDI) by multinational corporations (MNCs) have grown rapidly in the last two decades. For emerging market economies (EMEs) in particular, the number of CBMA, mostly by firms from developed markets, grew at an average annual rate of 14.5% during 1990-2014. While the determinants of the volumes of these flows are well studied, relatively little is known about what drives MNC ownership structure choices when acquiring EME firms. Yet, existing research has established that the extent of foreign ownership is an important determinant of a number of outcomes that have traditionally motivated policy makers to encourage FDI.  These include post-investment changes in labor productivity, wages, or export participation, and spillovers through technology transfer to subsidiaries.[1] In a forthcoming article, we set out to study the underlying determinants of FDI ownership structure in a broad set of EMEs.[2]

In a nutshell, our main argument is as follows.  Acquiring firms in EMEs entails both benefits and costs for MNCs from developed nations. Among the benefits, MNCs may have superior access to funding that they can use to relieve financial constraints of target firms, thus increasing the profitability of the acquired firm. At the same time, these acquisitions come with costs inflicted by weak local institutions, since operating firms in EMEs involves sourcing local inputs in an unfamiliar environment with insecure property rights and distortionary policies. So, how do MNCs deal with these competing forces? We show in our paper that an MNC’s choice of ownership structure is critical in balancing the aforementioned benefits and costs. To this end we develop a theoretical model that emphasizes the role of finance and institutions, and that delivers predictions about the optimal degree of foreign ownership, which we then take to the data.

To highlight the trade-offs facing a foreign acquirer, our theoretical model postulates that production in EMEs requires capital and a local input. The foreign acquirer solves for an optimal ownership contract between itself and the domestic target firm that captures its advantage in having greater access to capital markets relative to the credit-constrained target, and the potential disadvantages of operating a firm in an EME. The MNC’s disadvantage compared to local firms, which is due to weaker institutions, is modeled as a markup on local inputs that is paid only by an MNC. The markup thus incentivizes operating the firm with a local co-owner. The MNC then faces a choice between obtaining full control of the credit-constrained target, in which case it is compelled to pay a higher price for the local input, or to take partial ownership, in which case the domestic equity owner can provide the local input at a lower price.

Three distinctive sets of predictions emerge from the model. The first and second pertain to the ownership structure chosen by an MNC. Full (relative to partial) foreign ownership of targets is predicted to be more likely in sectors that have a greater dependence on external finance, and in countries that are less financially developed, while better institutions are found to tilt the scales towards full ownership. The effects of institutions and financial development are also predicted to be the largest for the sectors of the economy most dependent on external finance. The second set of predictions pertains to partial ownership. Here the model predicts that financial factors should play a weaker role in determining the precise size of partial stakes, while the input price markup is predicted not to influence the ownership structure in partial acquisitions at all. Our final predictions, which relate to the overall likelihood of foreign acquisitions, are that foreign acquisitions are more likely in sectors that have a greater dependence on external finance, in countries where financial markets are less developed, and when institutions are better.

We test these theoretical predictions in a large panel of CBMA transactions by developed market firms in fourteen EMEs over the period 1990-2007. We use the measure of sectoral external finance dependence due to Rajan and Zingales and country-level credit-GDP ratios as our main financial indicators, and anti-corruption indices as our baseline measure of institutional quality.[3]

The regression evidence confirms the main predictions of the model. The estimated effects are also quantitatively large. For example, the likelihood of an MNC choosing to own a domestic firm fully versus partially is predicted to be:

  • 22 percentage points larger for the sector with the highest (professional and scientific equipment) versus the lowest (tobacco) level of dependence on external finance
  • 21 percentage points lower in the most (Indonesia) versus the least (Chile) corrupt country
  • 14 percentage points lower in the most (South Africa) versus the least (Peru) financially developed country

As per the model, while dependence on external finance has the strongest effect in financially underdeveloped countries, it ceases to matter when local financial development, measured by private credit over GDP, exceeds 70%. In the same vein, external finance matters roughly three times more in countries with the lowest levels of corruption than in the most corrupt countries.

Our model’s predictions concerning the effect of financial factors on the overall prevalence of cross border acquisitions across sectors and countries are also borne out by the data. For example, we find that moving from the sector that is most to least dependent on external finance raises the share of CBMA (among all acquisitions) by 22 percentage points. At the same time, the share of CBMA is predicted to be 27 percentage points lower in the most versus the least financially developed country.

Taken together, our theoretical model and empirical evidence show that the interaction of financial, institutional, and technological factors plays an important role in determining the pattern of foreign ownership in North-South FDI flows. It also throws light on a number of empirical features of CBMA across sectors and countries, for example, why full foreign acquisitions are seldom observed (roughly 19%)  in countries such as Thailand that have both developed financial markets and weak institutions.  Our results also point towards improvements in institutions as a way to encourage higher MNC equity participation. For example, according to our estimates, a country like China would experience a doubling in the share of full acquisitions if it were to improve its corruption situation to the levels of Chile.


Alquist, R., N. Berman, R. Mukherjee, and L.L. Tesar, (forthcoming); “Financial Constraints, Institutions, and Foreign Ownership.” To appear in Journal of International Economics, DOI:

Bircan, Çağatay, (2019); “Ownership Structure and Productivity of Multinationals.” Journal of International Economics 116 (2019): 125-143.

Havranek, T., and Z. Irsova, (2011); “Estimating Vertical Spillovers from FDI: Why Results Vary and What the True Effect is.” Journal of International Economics 85(2): 234-244.

Javorcik, B. S., and M. Spatareanu, (2008); “To Share or Not to Share: Does Local Participation Matter for Spillovers from Foreign Direct Investment?Journal of Development Economics 85(1-2): 194-217.

Rajan, R., and L. Zingales, (1998); “Financial Dependence and Growth.” American Economic Review 88.3 (1998): 559-86.


[1] See for example, Javorcik and Spatareanu (2008),  Havranek and Irsova (2011), and Bircan (2019).

[2] See Alquist, Berman, Mukherjee and Tesar (forthcoming).

[3] See Rajan and Zingales (1998).

The WTO and Economic Development

Ben Zissimos (University of Exeter Business School)

My new edited volume tilted The WTO and Economic Development, brings together a collection of perspectives on different aspects of the purpose and institutional design of the World Trade Organization (WTO), and how these relate to economic development.[1]  The perspectives are contributed by a group of leading scholars in the economics of international trade.  The role that the WTO and its progenitor, the General Agreement on Tariffs and Trade (GATT), have played to date in facilitating economic development, and the role that the WTO can reasonably be expected to play in the future, is the unifying theme.

The following summary is based on my introductory chapter, which presents a synthetic literature review to develop context for the contributions that follow and draws basic insights.

Chapter 1, by Robert Staiger, sets out a comprehensive framework for formally incorporating non-tariff measures (NTMs) into a model for analyzing a multilateral trade agreement, taking tariffs into account as well.  The chapter notes that while developing countries tend to impose border NTMs on imports from developed countries, developed countries tend to impose behind-the-border NTMs on imports from developing countries.  A key contribution of the chapter is to show that an agreement involving border-NTMs is in fact amenable to a terms-of-trade motivation.  Since border-NTMs can exert a negative terms-of-trade externality on trade partners, by causing a reduction in demand for their exports, an agreement over border-NTMs has the same motivation of escaping from a terms-of-trade externality as in the conventional tariff-based ‘terms-of-trade theory’ of trade agreements.[2]

Chapter 2, by Chad Bown, adopts a more traditional focus on tariffs.  The motivation is compelling, arguing that there are 3.5 billion people in the world who have yet to benefit from an agreement to lower tariffs under the GATT/WTO, the overwhelming majority of whom are in developing countries.  The chapter tests for developing countries an implication of the terms-of-trade theory of trade agreements that has been shown to hold in developed countries.  The implication focused on in the chapter is that, through WTO negotiations, members are requested to take on lower tariff binding commitments in products for which they have higher market power, and thus where their tariffs (if left unchecked) would result in larger terms-of-trade externality losses for trade partners.  The chapter identifies well-defined groups of developing countries for which the implication holds, and groups for which it does not, showing that the terms-of-trade theory is relevant to developing-country trade liberalization through trade agreements but is not the only motivation.

Chapter 3, by Rodney Ludema, Anna Maria Mayda, and Jonathon McClure, studies the evolution of the so-called ‘MFN free rider problem’, an implication of the terms-of-trade theory.  In their earlier work, Ludema and Mayda show that an exporting country’s benefit from an MFN tariff concession by another country is proportional to exporter concentration.[3]  An exporting country’s willingness to pay for an MFN tariff concession on the product it exports with tariff concessions of its own depends on how much its refusal to offer concessions would reduce the MFN tariff concession.  The smaller the exporter, the less its refusal would mitigate the tariff cut, and thus the less costly it would be for the exporter to refuse to make a concession, thus free-riding on the concessions of other countries.  An intriguing contribution of the chapter is to show that, through the growth of trade with emerging economies such as China since 1993, the MFN free rider effect is found to have gotten worse.

Chapter 4, by Xuepeng Liu, considers a puzzle concerning so-called non-member participants (NMPs).  NMPs consist of three groups: colonies and overseas territories of GATT members; newly independent states; and provisional members.  NMPs are relevant here because they tend overwhelmingly to be developing countries.  The first econometric literature on the effects of the GATT/WTO explores whether member countries really have different trade patterns than outsiders, thus assessing the effectiveness of the GATT/WTO in liberalizing trade.[4]  The literature shows that they do, but in the process finds an ‘NMP puzzle’: while two formal GATT members trade 61 per cent more than the baseline case of neither country being a formal member nor an NMP, two NMPs trade 140 per cent more than the baseline.  It is counterintuitive that the NMPs should trade even more than formal members.  Chapter 4’s main contribution is to show that the ‘NMP puzzle’ can be resolved by undertaking two relatively simple modifications to the original gravity equation approach of the prior literature.

In Chapter 5, David DeRemer develops a model for analyzing a trade agreement when autarky is the (unique) outcome of non-cooperation over trade policy.  While the canonical model of trade agreements with perfect competition and political economy has proved to be powerful and flexible in explaining many aspects of trade liberalization under the GATT/WTO, it cannot motivate a trade agreement of the kind that DeRemer considers.[5]  Specifically, in the canonical model, if each government has a unilateral preference for autarky then they must have a joint preference for autarky as well.  This limits the scope for studying situations where developing countries have adopted autarkic trade policies for specific sectors, but where there may nevertheless be scope to open these sectors as part of a trade agreement.  For example, developing countries have commonly produced busses and trucks domestically behind high tariff walls.  The chapter adopts a familiar ‘Brander-Spencer’ type model in which to motivate and explore the scope for a trade agreement when autarky is the non-cooperative outcome.

Chapter 6, by Fabrice Defever and Alejandro Riaño, looks at the export promotion policies implemented by China, and how these have promoted the transition of China from autarky in the 1970s to the world’s largest exporting economy today.  The point of departure for this chapter is a set of stylized facts on firm exporting behavior that has been established in the economics literature for the world’s major trading economies: relatively few firms engage in exporting; exporting firms tend to be more productive and hence larger; most firms that do export sell only a small fraction of their output abroad.[6]  The chapter reveals that, on the face of it, the characteristics of Chinese exporters fit the stylized facts listed above. The most striking difference, the chapter finds, is that a third of firms export almost all of their output: China is thus characterized as having a ‘dual export sector’.  The overall conclusion of the chapter is that China’s export promotion policies have been responsible for creating its dual export sector, and have been instrumental in China becoming the world’s largest exporter.

Chapter 7, by Eric Bond, considers whether an efficient trade agreement should allow for gradual trade liberalization to mitigate adjustment costs.  Recent research has shown that the adjustment costs of moving productive resources between sectors in response to trade liberalization are significantly higher than previously thought.[7]  These costs are likely to be particularly high for developing countries, where adjustment is likely to involve geographical relocation between rural and urban settings.  The analytical approach taken in Chapter 7 is to examine the optimal liberalization path between two large countries, where workers face adjustment costs of moving between sectors.  The results show that if tariffs are the only policy instruments available, then developing countries should be allowed longer phase-in periods if their marginal costs of adjustment are higher than in developed countries.  Hence, the analysis shows that there may be a normative justification for so-called ‘special and differential treatment’ of developing countries.

Chapter 8, by Eric Bond and Kamal Saggi, contrasts the roles of price controls and compulsory licensing (CL) to improve consumer access to patented foreign products in developing countries.  While the Trade Related Aspects of Intellectual Property Rights (TRIPS) agreement created a storm of controversy, the eye of the storm was over the implication that, as a result of the agreement, it became more difficult for poor people in developing countries to access medicine at affordable prices.  Under the terms of the TRIPS agreement, if a patent holder refuses to grant access to its product on ‘reasonable’ commercial terms then a government may grant a CL to a different firm to produce the product.  The main lesson of the chapter is that the social value of CL depends crucially on entry costs and the size of the market, and is ambiguous.  This ambiguity seems to be a feature of outcomes under the TRIPS agreement more broadly, making it difficult to assess the extent to which it is beneficial or harmful overall.

The ninth and final chapter, by Mostafa Beshkar and Mahdi Majbouri, tests empirically the outcomes of disputes, focusing on whether or not they lead to litigation, taking explicit account of whether or not the dispute involves developed and/or developing countries.  The chapter focuses on the fact that developing and developed countries show divergent behavior in the dispute settlement process.  A surprising pattern uncovered in Chapter 9 is that, in a dispute between a developed and a developing country, litigation is more likely if the developed country is the defending party.  As detailed in the chapter, 62 per cent of disputes in which a developed country presses charges against a developing country are settled without establishing a dispute panel. In contrast, only 44 per cent of disputes are settled without establishing a dispute panel if a developing country mounts a dispute against a developed country.  Importantly, the chapter shows econometrically that this asymmetry disappears after 2001, when the Advisory Centre on WTO Law (ACWL) was established to make available advice and subsidies to poorer countries, to help them with the costs of mounting a WTO dispute.


Bagwell, K., and R.W. Staiger, (1999); “An Economic Theory of the GATT.”  American Economic Review 89: 215-248.

Bagwell, K., and R.W. Staiger, (2002); The Economics of the World Trading System.  MIT Press, Cambridge (Mass), US.

Bernard, A.B., J.B. Jensen, S.J. Redding, and P.K. Schott, (2007); “Firms in International Trade.” Journal of Economic Perspectives 21(3): 105-130.

Dix-Carneiro, R., (2014); “Trade Liberalization and Labor Market Dynamics.” Econometrica 82(3): 825-885.

Ludema, R., and A.M. Mayda, (2009); “Do Countries Free Ride on MFN?” Journal of International Economics 77(2): 137-150.

Ludema, R., and A.M. Mayda, (2013); “Do Terms-of-Trade Effects Matter for Trade Agreements? Theory and Evidence from WTO Countries.” Quarterly Journal of Economics 128(4): 1837- 1893.

Melitz, M.J., and S.J. Redding (2014); “Heterogeneous Firms and Trade.” Handbook of International Economics, 4th ed, 4: 1-54.

Rose, A., (2004); “Do We Really Know That the WTO Increases Trade?” American Economic Review 94(1): 98-114.

Tomz, M., J.L. Goldstein, and D. Rivers, (2007); “Do We Really Know That the WTO Increases Trade? Comment.”  American Economic Review 97(5): 2005-2018.

Zissimos, B., (forthcoming) The WTO and Economic Development, accepted for publication by MIT Press, Cambridge (Mass), US.


[1] See Zissimos (forthcoming).  The MIT Press have kindly allowed me to post the full text of this volume on my website until the book appears in print.  Please see the above reference for a link.

[2] See Bagwell and Staiger (1999, 2002).

[3] See Ludema and Mayda (2009, 2013).

[4] See Rose (2004), and Tomz, Goldstein and Rivers (2007).

[5] The canonical model is due to Bagwell and Staiger (1999, 2002).

[6] See Bernard, Jensen, Redding and Schott (2007), and Melitz and Redding (2014).

[7] See Dix-Carneiro (2014).