Nathaniel Beck

Throwing Out the Baby With the Bath Water: A Comment on Green, Kim and Yoon

Donald P. Green, Soo Yeon Kim, and David H. Yoon argue that many findings in quantitative international relations that use the dyad-year design are flawed. In particular, they argue that the effect of democracy on both trade and conflict has been vastly overstated, that researchers have ignored unobserved heterogeneity between the various dyads, and that heterogeneity can be best modeled by “fixed effects,” that is, a model that includes a separate dummy for each dyad.

Beyond linearity by default: Generalized additive models

Social scientists almost always use statistical models positing the dependent variable as a global, linear function of X, despite suspicions that the social and political world is not so simple, or that our theories are so strong. Generalized additive models (GAMs) let researchers fit each independent variable with arbitrary nonparametric functions, but subject to the constraint that the nonparametric effects combine additively. In this way GAMs strike a sensible balance between the flexibility of nonparametric techniques and the ease of interpretation and familiarity of linear regression. GAMs thus offer social scientists a practical methodology for improving on the extant practice of global linearity by default. We reanalyze published work from several subfields of political science, highlighting the strengths (and limitations) of GAMs. We estimate non-linear marginal effects in a regression analysis of incumbent reelection, nonparametric duration dependence in an analysis of cabinet duration, and within-dyad interaction effects in a reconsideration of the democratic peace hypothesis. We conclude with a more general consideration of the circumstances in which GAMs are likely to be of use to political scientists, as well as some apparent limitations of the technique.

Parties, Administrations, and American Macroeconomic Outcomes

This study re-examines Hibbss (1977b) findings on the impact of political party on unemployment rates for the postwar United States. With the Use of data through the end of the Carter administration and slightly different methods, it is concluded that party has between one-third and one-half the impact on unemployment claimed by Hibbs. Administration is a better predictor of unemployment than is party; in particular, neither the Kennedy nor Carter administration behaved as Hibbs claims Democratic administrations should, nor did the Nixon administration behave as Hibbs claims Republican administrations should. The significance of whether administration or party is the better predictor of economic outcomes is explored.

Presidential Influence on the Federal Reserve in the 1970’s

This research examined presidential influence on Federal Reserve monetary policy in the 1970s. A reaction function for the change in the Federal Funds rate was estimated and then checked to determine whether that reaction function shifted at politically important points. The two political influences studied were elections and administration. The Federal Reserve did not appear to manipulate the money supply in 1972 to aid President Nixons reelection. The reaction function shifted between the Nixon and Ford administrations; it did not shift with the ascension of Carter, and it did not shift with the replacement of Chairman Arthur Burns by Chairman William Miller.

Time‐series–cross‐section Data

This article treats the analysis of ‘time‐series–cross‐section’ (TSCS) data. Such data consists of repeated observations on a series of fixed units. Examples of such data are annual observations on the political economy of OECD nations in the post‐war era. TSCS data is distinguished from ‘panel’ data, in that asymptotics are in the number of repeated observations, not the number of units. The article begins by treating the complications of TSCS data in an ‘old‐fashioned’ manner, that is, as a nuisance which causes estimation difficulties. It claims that TSCS data should be analyzed via ordinary least squares with ‘panel correct standard errors’ rather than generalized least squares methods. Dynamics should be modeled via a lagged dependent variable or, if appropriate, a single equation error correction model. The article then treats more modern issues, in particular, the modeling of spatial effects and heterogeneity. It also claims that heterogeneity should be assessed with ‘panel cross‐validation’ as well as more standard tests. The article concludes with a discussion of estimation in the presence of a binary dependent variable.

Government Partisanship, Labor Organization, and Macroeconomic Performance: A Corrigendum

Alvarez, Garrett and Lange (1991) used cross-national data panel data on the Organization for Economic Coordination and Development nations to show that countries with left governments and encompassing labor movements enjoyed superior economic performance. Here we show that the standard errors reported in that article are incorrect. Reestimation of the model using ordinary least squares and robust standard errors upholds the major finding of Alvarez, Garrett and Lange, regarding the political and institutional causes of economic growth but leaves the findings for unemployment and inflation open to question. We show that the model used by Alvarez, Garrett and Lange, feasible generalized least squares, cannot produce standard errors when the number of countries analyzed exceeds the length of the time period under analysis. Also, we argue that ordinary least squares with robust standard errors is superior to feasible generalized least square for typical cross-national panel studies.

Domestic Political Sources of American Monetary Policy: 1955–82

This study investigates what domestic political factors affect monetary policy in the United States. Monetary policy is measured by changes in adjusted bank reserves. A reaction function is estimated using quarterly data for 1955-82. Independent economic variables in the reaction function are current and expected inflation, slackness, international reserves, the balance of payments, and the high employment surplus. Political variables examined are elections, party, administration, and the relationship between monetary and fiscal policy. Elections do not affect Fed policy. Monetary policy is easier under Democratic presidents; the Kennedy and Nixon administrations do not fit this general pattern. The effect of party and administration is linear; neither party nor administration affects the relationship between the state of the economy and Fed policy. Monetary and fiscal policy covaried during the 1960s; by the 1970s easy fiscal and tight monetary policy became more common.

Alternative Models of Dynamics in Binary Time-Series-Cross-Section Models: The Example of State Failure 1

This paper investigates a variety of dynamic probit models for time-series– cross-section data in the context of explaining state failure. It shows that ordinary probit, which ignores dynamics, is misleading. Alternatives that seem to produce sensible results are the transition model and a model which includes a lagged latent dependent variable. It is argued that the use of a lagged latent variable is often superior to the use of a lagged realized dependent variable. It is also shown that the latter is a special case of the transition model. The relationship between the transition model and event history methods is also considered: the transition model estimates an event history model for both values of the dependent variable, yielding estimates that are identical to those produced by the two event history models. Furthermore, one can incorporate the insights gleaned from the event history models into the transition analysis, so that researchers do not have to assume duration independence. The conclusion notes that investigations of the various models have been limited to data sets which contain long sequences of zeros; models may perform differently in data sets with shorter bursts of zeros and ones.

Theory and Evidence in International Conflict: A Response to De Marchi, Gelpi, and Grynaviski

In this article, we show that de Marchi, Gelpi, and Grynaviskis substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model.

Does there exist a political business cycle: A Box-Tiao analysis

Theories of the political business cycle are by now common. They range from the elegant theoretical work of Nordhaus (1975) to the essentially empirical work of Tufte (1978). They all assume that politicians manipulate the economy to aid their reelection efforts; in particular, politicians manipulate macroeconomic policy to create a boom just before election day. Golden and Poterba (1980) provide an excellent review of this literature. Both Tufte and Nordhaus provide casual evidence for the existence of a political business cycle. Using inflation and unemployment rates as indicators of macroeconomic outcomes, Tufte and Nordhaus predict at a minimum, that unemployment should be decreasing before election day. If presidents are really manipulating the economy very well, then unemployment should be at a minimum in the October before the election, and hence should be rising after the election. In the Nordhaus version, the economy pays for the pre-election boom with a post-election surge in the inflation rate. Nordhaus also suggests that the president should engineer a recession early in his term to reduce inflationary expectations. Tufte (1978: 20) presents graphs of the unemployment series and claims that visual examination shows them to be generally consistent with the hypothesis. However, he provides no statistical tests of the hypothesis, nor does he estimate the quantitative impact of elections on unemployment. Tufte does not examine the inflation series. Nordhaus restricts himself to comparing unemployment rates in the two years preceeding and following an election, with a simple nonparametric test supporting the hypothesis. The literature provides several good empirical tests of the political business cycle hypothesis. Both Paldam (1979) and McCallum (1978) test for the existence of four year cycles, and find none. However they do not test for discontinuous changes in the series around election day. Golden and Poterba examine whether the use of policy instruments changes near election day (and