Research - Pedro H. Albuquerque



Refereed Journal Articles

Articles Submitted to Refereed Journals

Working Papers

Trade Journal Articles

Book Chapters

Econometric Codes

My EconPapers (RePEc) Page

My SSRN Page

Refereed Journal Articles

The Effect of Oil Price on China’s Exports

With  João Ricardo Faria, André Varella Mollick and Miguel A. León-Ledesma.

China Economic Review (2009), 20 (4), 793-805.

Top 14 Hottest Article from October to December 2009.

The increase in oil prices in recent years has occurred concurrently with a rapid expansion of Chinese exports in the world markets, despite China being an oil importing country. In this paper we develop a theoretical model that explains the positive correlation between Chinese exports and the oil price. The model shows that Chinese growth can lead to an increase in oil prices that has a stronger impact on its export competitors. This is due to the large labor force surplus of China. We then examine this hypothesis by estimating a reduced form equation for Chinese exports using Rodrik (2006)’s measure of export competitiveness, together with the oil price, productivity, real exchange rate, and foreign industrial production over the monthly 1992-2005 period. The results suggest a stable relationship and yields slightly positive values for the price of oil and elastic coefficients for export competitiveness, along with the expected negative elasticity for the real exchange rate.

Canaries and Vultures: A Quantitative History of Monetary Mismanagement in Brazil

With Solange Gouvea.

Journal of International Money and Finance (2009), 28 (3), 479-495.

During the last two decades of the twentieth century, Brazil went through a sequence of failed stabilization plans that tried to cope with an enduring hyperinflation. This paper uses a money demand model to evaluate monetary policies during those episodes. Consistency between money supply and expected conditional money demand growth rates is considered for each plan. It is shown that unsuccessful programs were marked by excessive liquidity. The results not only suggest that monetary mismanagement led to the failure of the plans, but also that the excessive liquidity could have been predicted.

Can Globalisation Stop the Decline in Commodities’ Terms of Trade?

With André Varella Mollick, João Ricardo Faria and Miguel A. León-Ledesma.

Cambridge Journal of Economics (2008), 32 (5), 683-701.

Fifth Most-Frequently Read Article in October 2008.

In this paper we address the following question: would a fully integrated world economy eliminate the widely reported decline in the terms of trade of primary commodities? We address the question by looking at the terms of trade within the US (a highly integrated economy). Our findings show two results. First, US internal real commodities' terms of trade over the 1947–1998 period experienced slowly declining but significant trends. Second, once we control for the effect of US prices on international terms of trade, we find a long-run relationship between the US and international relative prices. These findings support the view that the decline of commodities' terms of trade bears no relationship with the process of globalisation. This seems to indicate that, if world terms of trade behaved as the US terms of trade, neither increased integration nor protectionist measures would eliminate this trend.

Shared Legacies, Disparate Outcomes: Why American South Border Cities Turned the Tables on Crime and Their Mexican Sisters Did Not

Crime, Law and Social Change (2007), 47 (2), 69-88.

The article evaluates crime trends in south border American and Mexican sister cities using panel data analysis. The region offers a unique assessment opportunity since cities are characterized by shared cultural and historical legacies, institutional heterogeneity, and disparate crime outcomes. Higher homicide rates on the Mexican side seem to result from deficient law enforcement. Higher population densities in Mexican cities appear to also be a factor. Cultural differences, on the other hand, have been decreasing, and apparently do not play a substantial role. The homicide rate dynamics show opportunistic clustering of criminal activity in Mexican cities, while no clustering is found on the American side. Crime also appears to spill from Mexican cities into American cities. Homicide rates on both sides of the border have been falling faster than countrywide rates, leading, in the case of American cities, and against stereotypes, to rates below the countrywide rate in 2001.


See also the trade journal article "How Laredo Turned the Tables on Crime"

BAD Taxation: Disintermediation and Illiquidity in a Bank Account Debits Tax Model

International Tax and Public Finance (2006), 13 (5), 601-624.

This paper uses a dynamic general equilibrium model to study the economic effects of bank account debits (BAD) taxation. Australia and various Latin American countries have levied or levy BAD taxes. Aspects such as financial disintermediation, market illiquidity, and impacts on dividend and interest rates are considered. Part of the BAD tax revenue may be fictitious, due to increased interest payments on government debt. The Brazilian BAD tax (CPMF) experience is evaluated. The empirical analysis confirms some theoretical predictions. Incidence base over GDP appears to be sensitive to the tax rate, possibly engendering a Laffer curve. The tax may also cause real interest rates to increase. Furthermore, the deadweight losses are relatively large, even if revenues are small. The theoretical and empirical results suggest that the BAD tax is not adequate for revenue collection.

Are European Stock Markets Influencing Latin American Stock Markets?

With Andrés Rivas and Antonio Rodriguez.

Análisis Económico (2006), 21 (47), 51-67.

In this study, we examine the response of Latin American stock markets to movements in European stock markets. Our results vary depending on the openness of the country in terms of international trade. We find evidence that Latin American stock markets are affected by Spanish stock market. Additionally, during the second and third-periods (1995 to 1998 and 1999 to 2004) Spain appears to have much stronger ties (such as more trade) with Brazil and Chile, and this might explain why Brazil and Chile are affected from Spain and not from the other European markets. This study uncovers two important findings. First, Spain has an effect on Latin American markets but these responses are not homogeneous across markets. Second, the magnitude of Spain’s influence is different in each of the three sub-periods under study.

A Practical Log-Linear Aggregation Method with Examples: Heterogeneous Income Growth in the USA

Journal of Applied Econometrics (2003), 18 (6), 665-678.

Journal of Applied Econometrics Data Archive

A practical aggregation method for heterogeneous log-linear functions is presented. Inequality measures are employed in the construction of a simple but exact aggregate representation of an economy. Three macroeconomic applications are discussed: the aggregation of the Lucas supply function, the time-inconsistent behaviour of an egalitarian social planner facing heterogeneous discount rates, and the case of a simple heterogeneous growth model. In the latter application, aggregate CPS data is used to show that the slowdown that followed the first oil shock is worse than usually thought, and that the "new economy" growth resurgence is not as strong as it appears.

Articles Submitted to Refereed Journals

Optimal Time Interval Selection in Long-Run Correlation Estimation

This paper presents an asymptotically optimal time interval selection criterion for the long-run correlation block estimator (Bartlett kernel estimator) based on the Newey-West approach. An alignment criterion that enhances finite-sample performance is also proposed. The procedure offers an optimal yet unobtrusive alternative to the common practice in finance and economics of arbitrarily choosing time intervals or lags in correlation studies. A Monte Carlo experiment using parameters derived from Dow Jones returns data confirms that the procedure is MSE-superior to typical alternatives such as aggregation over arbitrary time intervals, VAR estimation, and Newey-West automatic lag selection.

Femicide Rates in Mexican Cities along the US-Mexico Border: Do the Maquiladora Industries Play a Role?

With Prasad R. Vemala.

Mexican cities along the US-Mexico border, especially Cd. Juarez, became notorious for supposedly presenting abnormally high femicide rates which many believe to be associated with the development of maquiladora industries in the region, partially as a result of the NAFTA. Nonetheless, statistical evaluation of data ranging from 1990 to 2012 shows that, after factoring in the effects of maquiladora employment and male homicide rates, femicide rates in the region, including Cd. Juarez, are consistent with rates in non-border Mexican cities where maquiladora industries are relatively less present, and tend to fall with increased employment opportunities in maquiladora industries. Femicide rates in Cd. Juarez, although high, are in most years similar to rates in Houston and Nuevo Laredo and, as a share of overall homicide rates, are lower than in most other cities considered in the study. These results challenge conventional wisdom and most of the existing literature on the US-Mexico border femicide phenomenon.

Financial Stability, Natural Disasters, and Political Conflicts: Time Series Evidence on Causality in Developing Countries

With Wassim Rajhi.

We adopt a panel structural VAR modeling approach to model the causal effects of natural disasters and state fragility on aggregate financial system variables such as deposits, non-performing loans, and GDP per capita. We divide the sample by banking sector variable and we develop 8 panels across more than 63 countries over a 13 year period (1999-2011) for all countries in the dataset, excluding high-income countries. It is noteworthy that these analyses are the first of their kind to evaluate empirically the link between natural disasters, state fragility and banking variables. Our main findings suggest that lifeyears lost index can increase the amount of aggregate non-performing loans and the bank's z-score up to two to five years after the shock. Finally, State fragility increases the non-performing loans and the likelihood of bank default in developing countries, as proxied by their z-score which depletes the banks’ reserves and increases the leverage. Our results also indicate that natural disasters and state fragility decrease the banking deposits and financial system deposits.

Technological Innovations, Remanufacturing and Green Accounting

This article uses a vintage capital theoretical framework to investigate how technological innovations and remanufacturing can affect national and green accounting differently, and how green (Pigouvian) taxes can promote remanufacturing and improve social welfare, in agreement with circular economy and sustainability principles. Technological innovations embodied in new capital vintages lead to increases in productivity but also to natural resources extraction and waste when capital of new vintages is produced and capital of older vintages is discarded. Remanufacturing can reduce social costs of innovations by allowing embodiment to happen with less resources extraction and waste, but it can increase private costs of production and decrease productivity, so the trade-offs between social and private costs need to be evaluated. National accounting measurements of economic output do not fully consider social costs, while green accounting measurements, when correctly defined, are more representative of social welfare. Modeling the phenomenon is important because it allows for better understanding of the effects of remanufacturing on economic output and social welfare, and because it offers conceptual guidelines for the development of systems of green accounting and for the design of green taxes.

Working Papers

Inequality-Driven Growth: Unveiling Aggregation Effects in Growth Equations

It is well known from nonlinear aggregation theory that distributions play a central role in the determination of the values of aggregate variables. This paper establishes a bridge between the aggregation and the inequality and growth literature by applying a log-linear aggregation method to a simple Cass-Koopmans-Ramsey AK growth model with heterogeneity. The aggregation effect is explicitly captured in the aggregate growth rate equation by the changes of the mean logarithmic deviation (MLD or Theil’s second measure) of the income, implying that increases in income inequality may be unambiguously associated with temporary increases of a country’s growth rate, in agreement with the empirical findings of Forbes (AER, 2000). Consequently, empirical studies of the long-run effects of income inequality may suffer from aggregation bias if the temporary effects of the MLD changes are not considered. The accelerated growth episodes observed in Brazil from 1968 to 1973 and in China recently demonstrate that the increase in income inequality, as measured by the MLD changes, may have resulted, through the aggregation effect, in substantial temporary increases of the aggregate growth rates experienced by those countries.

Book Chapters

Os Impactos Econômicos da CPMF: Teoria e Evidência

(The Economic Impacts of the CPMF: Theory and Evidence)

Winner of the 2nd prize in the VI Brazilian National Treasury Competition, Topics in Public Finance, published in Finanças Públicas: VI Prêmio Tesouro Nacional – 2001. STN: Brasília.

Este trabalho tem como objetivo estudar os impactos econômicos da CPMF na economia brasileira, e para isto está dividido em três blocos. No primeiro a CPMF é analisada sob a ótica da teoria econômica. No segundo é discutida a experiência internacional com impostos similares à CPMF. No terceiro a experiência brasileira é examinada com o auxílio dos dois blocos anteriores. Segundo a teoria econômica, a CPMF causaria a elevação dos juros reais de modo desproporcional a outros impostos, o que seria o resultado da inclusão da rotatividade de ativos em sua base de incidência. Esta deficiência em sua concepção afetaria negativamente, e de forma desproporcional à sua arrecadação, os níveis de capital, produção e salários. Ela também causaria o aumento das despesas do governo com pagamento de juros, o que levaria parte de sua receita a ser fictícia. A CPMF causaria desintermediação e iliquidez nos mercados financeiros, desincentivando o ressurgimento do crédito. A arrecadação comportar-se-ia de acordo com uma curva de Laffer, com elevadas perdas de peso morto, particularmente quando comparadas à pequena receita resultante. Resultados empíricos confirmam que tais conclusões seriam aplicáveis ao caso brasileiro. A teoria econômica, a experiência internacional e a evidência brasileira revelam, portanto, que a CPMF apresenta significativas deficiências como instrumento de arrecadação.

(This paper aims to study the economic impacts of the CPMF in the Brazilian economy and is divided in three parts. In the first part, the CPMF is analyzed under the scope of economic theory. In the second part, the international experience with taxes similar to the CPMF (BAD taxes) is discussed. In the third part, the Brazilian experience is analyzed, considered the contributions of the two previous parts. According to the economic theory, the CPMF should increase real interest rates unproportionally to other taxes, because of the inclusion of asset turnovers in its incidence base. This conceptual deficiency would negatively affect, unproportionally to its tax collection, the level of capital, production and wages. It would also cause the increase of government spending through interest payments, implying that a part of the revenue is fictitious. The CPMF would cause disintermediation and illiquidity in the financial markets, hindering the resurgence of credit in Brazil. The tax collection would follow a Laffer curve, with high deadweight losses, particularly when compared to its small revenue. Empirical results confirm that those conclusions may be valid in the Brazilian case. The economic theory, the international experience and the Brazilian evidence reveal, therefore, that the CPMF presents significant deficiencies as a tax collection instrument.)

Econometric Codes

Long-Run Correlation Estimator

Code for EViews

Description: this code for EViews calculates pairwise nonparametric long-run correlation (coherency at frequency zero) estimates using optimal time interval selection and alignment criteria, as described in "Optimal Time Interval Selection in Long-Run Correlation Estimation."

Instructions: input the following parameters in the EViews "run" parameter window: minimum alignment, maximum alignment, sample start date, sample end date, and the list of variables between quotation marks. The output table name is "lrcorr."

Note: make sure that the sample does not contain NA observations, otherwise the code may not run.

Parameter input example:

 -7 7 1/02/1990 12/30/2001 "dldow dlibovespa dlipc dlmerval dlnasdaq"

where dlnasdaq, for instance, represents the NASDAQ daily rate of return.

Long-Run Correlation Estimator Code for Excel


Description: this worksheet for Excel calculates nonparametric long-run correlation (coherency at frequency zero) estimates between two series,  using the optimal time interval selection and alignment criteria described in "Optimal Time Interval Selection in Long-Run Correlation Estimation." Just unzip the file and follow the instructions on the worksheet. Enjoy!