Timothy G. Conley

GMM estimation with cross sectional dependence

This paper presents a spatial model of dependence among agents using a metric of economic distance. Measurements of this economic distance provide cross-sectional data with a structure similar to that provided by the time index in time-series data. Generalized method of moments estimators using such dependent data are shown to be consistent and asymptotically normal. This paper presents a class of non-parametric, positive semi-definite covariance matrix estimators that allow for general forms of dependence characterized by economic distance. These covariance matrix estimators are shown to remain consistent when economic distances are not precisely observed.

A Spatial Analysis of Sectoral Complementarity

This paper presents a spatial econometric method for characterizing productivity comovement across sectors of the U.S. economy. Input‐output relations provide an economic distance measure that is used to characterize interactions between sectors, as well as conduct estimation and inference. We construct two different economic distance measures. One metric implies that two sectors are close to one another if they use inputs of other industrial sectors in nearly the same proportion, and the other metric implies that sectors are close if their outputs are used by the same sectors. Our model holds that covariance in productivity growth across sectors is a function of economic distance. We find that (1) positive cross‐sector covariance of productivity growth generates a substantial fraction of the variance in aggregate productivity, (2) cross‐sector productivity covariance tends to be greatest between sectors with similar input relations, and (3) there are constant to modest increasing returns to scale. We test and reject the hypothesis that these correlations are due to a common shock.

A new semiparametric spatial model for panel time series

Abstract This paper presents a semiparametric model for large-dimension vector time series whose elements correspond to economic agents. Dependence between agents’ variables is characterized using a spatial model. Functions of agents’ economic distances provide restrictions that enable estimation of a vector autoregressive specification. We present sufficient conditions for our model to generate stationary, β-mixing series with finite higher-order moments. We estimate the model using a simple two-step sieve least-squares procedure, where the sieve estimators are constructed to preserve shape restrictions on the functions of economic distance, e.g., positive definiteness of a covariance function. We provide rates of convergence for the sieve estimators, T limiting distributions for the models finite-dimensional parameters, and a bootstrap method for inference. In an illustrative application, we use this model to characterize how the comovement in output growth across US industrial sectors depends on the similarity of sectors’ technologies. We also present a small Monte Carlo evaluation of our estimators.

Nativity and Wealth in Mid-Nineteenth-Century Cities

This article uses evidence from the manuscripts of the 1860 federal census to analyze the wealth of adult males in Boston, New York, Chicago, and Indianapolis. Previous multivariate analyses of wealth from the census have been flawed by reliance on ordinary least squares; we instead use quantile regression. Immigrants fared considerably better in the Midwest than the East: immigrants in the midwestern cities held more wealth than their eastern counterparts, both absolutely and relative to the native-born in their respective cities. We explore the causes of these differences and their consequences for nineteenth-century Americans and their communities.

The Fed Response to Equity Prices and Inflation

A number of researchers and market observers hold that the dramatic increase during the 1990s and subsequent decline in U.S. stock prices were due to non-fundamental factors, such as irrational expectations or bubbles. If this view is correct, policymakers may be concerned with the real macroeconomic consequences of the stock market run-up. These might include overconsumption due to a perceived wealth effect or too much physical investment due to a lower financing cost of capital. Following this reasoning, the Federal Reserve could raise the Fed Funds target rate to offset perceived non-fundamental stock price increases. This policy stance may seem particularly appealing if the Feds primary target, low and stable inflation, is already being achieved. This paper studies how Federal Reserve interestrate policy, from 1979:4 onward, responds to an aggregate measure of stock-market activity under high versus low inflation. Most existing research makes no distinction between policy across the highand low-inflation times of the past 24 years. Two conventional findings of this existing research are that the Federal Reserve: (i) raises the short-term real interest rate in response to inflation and (ii) does not change policy in response to equity price movements.1

Spillovers from Local Market Human Capital and the Spatial Distribution of Productivity in Malaysia

Abstract This paper examines whether spillovers from local market human capital are important in explaining the distribution of productivity across Malaysia. We develop an empirical method for describing local human capital distributions based on the idea that spillovers are limited in scope by costs of interaction or economic distance between agents. We use estimates of the economic distance between agents to construct measures of local market human capital based on schooling rates of the population within a given radius. These measures are then used in estimating equations obtained from a simple local public goods model. Our regressions are estimated using spatial GMM, allowing for general spatial correlation across observations as a function of economic distance. We find positive wage and rent differentials associated with local human capital, evidence consistent with productive human capital spillovers. Our results for rent differentials obtain with two distinct human capital measures; however, those for wage differentials depend on the human capital measure used.

BOOTSTRAPPING THE LONG RUN

We develop and apply bootstrap methods for diffusion models when fitted to the long run as characterized by the stationary distribution of the data. To obtain bootstrap refinements to statistical inference, we simulate candidate diffusion processes. We use these bootstrap methods to assess measurements of local mean reversion or “pull†to the center of the distribution for short-term interest rates. We also use them to evaluate the fit of the model to the empirical density.