You are asked to further reflect and critically elaborate on one of the lecture topics covered in class by reviewing a published research paper from the list of recommended readings under each topic.

You can follow the following outline for your essay (this is just a suggestion. You are welcome to come up with your own structure.)

• Background: Situate the paper within the broader lecture topic

• Theory: State the paper’s main argument

• Empirics: How is the argument being tested?

• Results: What is the paper’s main finding?

• Reflection/Discussion: What are the strengths/weaknesses of the paper? What questions arise from the reading? How could the paper be improved? Provide idea/s for future research that may arise from the paper.

The emphasis is on critical reflection. Simply repeating the arguments of the article is not enough.

Thus, the main part of the essay is the Reflection/Discussion part where you demonstrate that you are critically engaging with the paper and are embedding it in the broader topic that the paper is about (i.e., the lecture topic).

Instead, you should indicate where the author/s is/are wrong or where you see other perspectives the author should have given. You should point out the blind spots, prepare and raise up to 2 questions that emerge from reading the article.

Explaining Corruption: Are Open Countries Less Corrupt?

Introduction

Aside from the burden of paperwork and bureaucracy involved, they [exchange controls] are—surprise!—subject to abuse: Exporters have an incentive to hide their foreign-exchange receipts; importers, an incentive to pad their invoices.

Fighting corruption has progressively become an important item in many governments’ political agendas, as the adverse effects of corruption, have been widely recognized in policymakers’ discussions as well as in academic fora. For example, Mauro shows that corruption leads to lower growth rates (Mauro, 1995) and is associated with distortions in the composition of government expenditure (Mauro, 1998). Moreover, Kaufman et al. (1998) show that corruption has highly undesirable distributional impacts since it
disproportionately hurt.

The theoretical literature and anecdotal evidence point to a number of possible causes of corruption: low pay of public officials, inadequate systems of controls and sanctions within public administration, low public opinion pressure, cultural factors, and ethnic ties, and a high level of distortions in the economy (see the discussion in Alfiler, 1986).

In this work, I focus on a specific source of economic distortion—the presence of barriers to international trade and international capital flows.

2. SPECIFICATION, DATA, AND METHODOLOGY

To assess the impact of trade and financial openness, I estimate a basic model and then add in turn the variables that are meant to capture the different concepts of openness.

The core regressors are chosen so as to provide a parsimonious specification and, at the same time, to minimize the chances of omitted variable bias on the relevant coefficients. The specification is patterned along those of Ades and di Tella (1997) and Fisman and Gatti (2002).

Throughout the exercise, I control for the level of development of each country as proxied by (log) GDP per capita. Existing evidence indicates that poorer countries suffer from widespread corruption while at the same time they resort more often to trade taxation as a means to replenish government revenues (Easterly and Rebelo, 1993).

The basic model also includes an index measuring the extent of democracy to account for the fact that regular and open elections might provide a check on corruption and, at the same time, produce systematic patterns in trade policy choices. I also control for the size of the country, as proxied by (the log of) the population.

If large countries heavily exploit economies of scale in the provision of public services and therefore have a low ratio of public service outlets per population, individuals might resort to bribes ‘to get ahead of the queue’.2 At the same time, we know that small countries tend to be more open.

Therefore, controlling for the size of the countries is a priori important to eliminate a source of potential omitted variable bias on the coefficient of the openness variable. Finally, a time the trend is included.

3 EMPIRICAL EVIDENCE

3.1 Import Shares and Trade Tariffs

Table 2 reports results from the estimation of the basic model to which the trade openness variables are added individually and jointly (columns 1–6), and some robustness checks.

the coefficients on the core controls suggest some interesting regularities. Higher levels of development are associated—everything else being equal—with lower levels of corruption. The democracy index is negatively associated with corruption, suggesting that open and free elections might contribute to keeping corruption in check.

Consistent with the evidence in Ades and di Tella (1999), the import share enters the regression with a positive and significant coefficient (column 1). However, an analysis of the residuals suggests that the strength of the relationship is driven by the observations for Singapore, an extremely open country with virtually no reported corruption.6 When Singapore is removed, the association between import share and corruption is substantially weakened (column 2).

3.2 Non-Tariff Barriers

Favoring tariffs over non-tariff barriers is a textbook policy recommendation. Non-tariff barriers, when binding, generate rents which, in turn, might encourage corrupt transactions. To test whether this is the case, I add the percentage of goods subject to quotas to the basic regression (Table 3).

The non-tariff barrier measure does not enter the regression significantly, irrespective of whether import share and tariff rates are included in the specification. To the extent that quotas restrictions are not binding, we would not expect a significant association between quotas and corruption to emerge.
The estimated coefficients on trade and tariff rates confirm the results of the previous section, although here they emerge with lower precision, most likely due to the reduced sample size.

3.3 Financial Openness

We now investigate the association between financial openness and corruption (Table 4). None of the IMF dummies for financial markets restrictions is significant when entered in the regression individually (column 1 and 2), jointly (column 4), and together with the share of FDI (column 5).

One might attribute this lack of correlation to the fact that these indicators capture only the presence of controls and not their intensity. However, Quinn’s measure which is meant to measure the intensity of controls is also insignificant (column 3). The share of FDI shows a weak association with lower corruption, but analysis of the residuals suggests that it is driven by outliers