International Comparisons of Income, Prices, and Production

May 27-28, 2016
Robert Feenstra of the University of California, Davis and Alberto Cavallo of MIT, Organizers

Session I: Poverty and Cost of Living

Robert C. Allen, New York University Abu Dhabi

Absolute Poverty: When Necessity Displaces Desire

Allen proposes a new method for defining an international poverty line based on explicit budgeting. The novel feature of this paper is that linear programming is used to deduce the diet that minimizes cost and guarantees survival. Nonfood items are also explicitly budgeted and amount to about one quarter of the cost of subsistence. A series of least cost diets are calculated with increasingly demanding nutritional requirements for twenty countries using prices from ICP 2011. The aim is to see which requirements rationalize the spending pattern of the poor. The 'reduced basic' model does the job. When the cost of the nonfood items are added to the cost of the 'reduced basic' diet, the resulting Linear Programming Poverty Line (LPPL) averages $1.88 per day across the poor countries in the sample. The same model rationalizes both the spending pattern of the poor and the World Bank Poverty Line. The LPPL has the advantages that it is (1) clearly related to survival and well being, (2) comparable across time and space since the same nutritional requirements are used everywhere, (3) adjusts consumption patterns to local prices, (4) presents no index number problems since solutions are always in local prices, and (5) requires only readily available information, namely, the prices in ICP or equivalent.


Stephen J. Redding, Princeton University and NBER, and David Weinstein, Columbia University and NBER

A Unified Approach to Estimating Demand and Welfare

The measurement of price changes, economic welfare, and demand parameters is currently based on three disjoint approaches: macroeconomic models derived from time-invariant utility functions, microeconomic estimation based on time-varying utility (demand) systems, and actual price and real output data constructed using formulas that differ from either approach. The inconsistencies are so deep that the same assumptions that form the foundation of demand-system estimation can be used to prove that standard price indexes are incorrect, and the assumptions underlying standard exact and superlative price indexes invalidate demand-system estimation. In other words, Redding and Weinstein show that extant micro and macro welfare estimates are biased and inconsistent with each other as well as the data. They develop a unified approach to demand and price measurement that exactly rationalizes observed micro data on prices and expenditure shares while permitting exact aggregation and meaningful macro comparisons of welfare over time. They show that all standard price indexes are special cases of their approach for particular values of the elasticity of substitution, constant demand for each good, and a constant set of goods. In contrast to these standard index numbers, the researchers' approach allows them to compute changes in the cost of living that take into account both changes in the demand for individual goods and the entry and exit of goods over time. Using barcode data for the U.S. consumer goods industry, they show that allowing for the entry and exit of products, changing demand for individual goods, and a value for the elasticity of substitution estimated from the data yields substantially different conclusions for changes in the cost of living from standard index number.


Ingvild Almås, NHH; Timothy Beatty, University of California, Davis; and Thomas Crossley, Essex University

Lost in Translation: What do Engel Curves Tell Us About the Cost of Living?

The Hamilton method for estimating CPI bias is simple and intuitively appealing and has been widely adopted. Almås, Beatty, and Crossley show that the Hamilton method is internally inconsistent and conflates CPI bias with variation in cost of living across income levels. They demonstrate a method by which these two components can be disentangled. When applied to Hamilton's data the researchers' method gives larger estimates of CPI bias. Their approach can recover changes in cost of living for any income level and they exploit this to study growth in the median income for the United States.


Session II: Prices Within and Between Countries

Yuriy Gorodnichenko, University of California at Berkeley and NBER; Viacheslav Sheremirov, Federal Reserve Bank of Boston; and Oleksandr Talavera, University of Sheffield

Price Setting in Online Markets: Does IT Click? (NBER Working Paper No. 20819)

Using a unique dataset of daily U.S. and U.K. price listings and the associated number of clicks for precisely defined goods from a major shopping platform, Gorodnichenko, Sheremirov, and Talavera shed new light on how prices are set in online markets, which have a number of special properties such as low search costs, low costs of monitoring competitors' prices, and low costs of nominal price adjustment. The researchers document that although online prices change more frequently than offline prices, they nevertheless exhibit relatively long spells of fixed prices. By many metrics, such as large size and low synchronization of price changes, considerable cross-sectional dispersion, and low sensitivity to predictable or unanticipated changes in demand conditions, online prices are as imperfect as offline prices. The findings here suggest a need for more research on the sources of price rigidities and dispersion, as well as on the relative role of menu and search costs in online-pricing frictions.


Robert C. Feenstra; Mingzhi Xu, University of California at Davis; Alexis Antoniades, Georgetown University; and John Romalis, The University of Sydney and NBER

What is the Price of Tea in China? Towards the Relative Cost of Living in Chinese and U.S. Cities

Feenstra, Xu, Antoniades, and Romalis examine the price and variety of products at the barcode level in cities within China and the United States. In both countries, there is a greater variety of products in larger cities. But in China, unlike the United States, the prices of products tend to be lower in larger cities. The researchers attribute the lower prices to a pro-competitive effect, whereby large cities attract more firms which leads to lower markups and prices. Combining the effect of greater variety and lower prices, it follows that the cost-of-living for grocery-store products in China is lower in larger cities. They also compare the cost-of-living indexes for particular products between China and the United States. In most cases, the observed prices differences between the countries (lower prices in China) are partially offset by the variety differences (reduced variety in China), so that the cost of living in China is not as low as the price differences suggest, especially in smaller cities.

Brent Neiman, University of Chicago and NBER; Jonathan Eaton, Pennsylvania State University and NBER; and Samuel Kortum, Yale University and NBER

Obstfeld and Rogoff's International Macro Puzzles: A Quantitative Assessment (NBER Working Paper No. 21774)

Obstfeld and Rogoff (2001) propose that trade frictions lie behind key puzzles in international macroeconomics. Eaton, Kortum, and Neiman take a dynamic multicountry model of international trade, production, and investment to data from 19 countries to assess this proposition quantitatively. Using the framework developed in Eaton, Kortum, Neiman, and Romalis (2016), they revisit the puzzles in a counterfactual world without trade frictions in manufactures. Removing these trade frictions goes a long way toward resolving a number of the puzzles: The dependence of domestic investment on domestic saving falls by half or disappears entirely, mitigating the Feldstein-Horioka (1980) puzzle. Changes in nominal GDPs in U.S. dollars become less variable across countries and line up with changes in real GDPs as much as with real exchange rates, mitigating the exchange rate disconnect puzzle. Less dramatically, changes in consumption become more correlated across countries, mitigating the consumption correlations puzzle and changes in real exchange rates become less variable across countries, mitigating the relative purchasing power parity puzzle.


Session III: Output and Capital

Gholamreza Hajargasht, University of Melbourne, and Prasada Rao, University of Queensland

Computation of Standard Errors for Purchasing Power Parity Exchange Rates from the International Comparison Program


Mario J. Crucini, Vanderbilt University and NBER

Globalization Boom and Bust: A Study of US Automobile Exports from 1913 to 1940


Robert C. Feenstra, and Robert Inklaar and Marcel Timmer, University of Groningen

Penn World Table: version 9.0 and beyond


Session IV: Methods for Prices, Productivity, and Housing

Ina Simonovska, University of California at Davis and NBER; Joel David, University of Southern California; and Espen Henriksen, University of California at Davis

The Risky Capital of Emerging Markets (NBER Working Paper No. 20769)

Emerging markets exhibit high returns to capital, the 'Lucas Paradox,' alongside volatile growth rate regimes. David, Henriksen, and Simonovska investigate the role of long-run risks, i.e., risk due to fluctuations in economic growth rates, in leading to return differentials across countries. They take the perspective of a U.S. investor and outline an empirical strategy to identify risky growth shocks and quantify their implications. Long-run risks account for 60-70% of the observed return disparity between the U.S. and a group of the poorest countries. At the individual country level, the researchers’ model predicts average returns that are highly correlated with those in the data (0.61).


W. Erwin Diewert and Robert Inklaar, University of British Columbia and NBER

Measuring Industry Productivity across Time and Space and Cross Country Convergence

Inklaar and Diewert introduce a new method for simultaneously comparing industry productivity levels across countries and over time. The new method is similar to the method for making multilateral comparisons of Caves, Christensen and Diewert (1982b) but their method can only compare gross outputs across production units and not compare real value added of production units across time and space. The present paper uses the translog GDP methodology for measuring productivity levels across time that was pioneered by Diewert and Morrison (1986) and adapts it to the multilateral context. The new method is illustrated using an industry level data set and shows that productivity dispersion across 38 countries between 1995 and 2011 has decreased faster in the traded sector than in the non-traded sector. In both sectors, there is little evidence of decreasing distance to the productivity frontier.


Robert J. Hill, Miriam Steurer, Sofie Waltl, and Michael Scholz, University of Graz

The Treatment of Owner Occupied Housing in the CPI and its Implications for Monetary Policy

The impact of movements in house prices on the CPI depends on how owner-occupied housing (OOH) is included. Most national statistical institutes (NSIs) measure the cost of OOH using either the acquisitions or rental equivalence approaches. Hill, Steurer, Waltl, and Scholz argue here that the user cost approach is potentially better than either acquisitions or rental equivalence. However, the performance of the user cost method depends critically on how capital gains (actual or expected) are treated. From an axiomatic perspective the researchers argue that a case can be made for excluding capital gains. This also makes the CPI more responsive to the housing market, which may be desirable from a monetary policy perspective. Using detailed micro data for Sydney, Australia, the researchers then compare empirically the impact of these approaches on the CPI. Their results indicate that the CPI is very sensitive to the way OOH costs are measured. In the case of Sydney with its booming housing market, however, the user cost method with capital gains excluded pushes up the CPI so much as to undermine the feasibility of this approach. A user cost approach that extrapolates expected real capital gains over a long time horizon of about 30 years therefore may be preferable. These findings have important implications for the debate over how monetary policy should respond to booms and busts in the housing market.