Innovations in Economics Measurement
Paper Session
Friday, Jan. 6, 2023 10:15 AM - 12:15 PM (CST)
- Chair: John Haltiwanger, University of Maryland
Flexible Entry/Exit Adjustment for Price Indices
Abstract
This paper introduces and implements a new method for constructing an entry/exit adjusted price index based on a flexible demand system and estimation strategy. Entry/exit adjustment requires estimating demand system parameters to recover price changes which are unobserved when goods enter and exit. Popular CES-based entry/exit adjustment imposes strong restrictions on the distribution of elasticities both across products and over time, but only requires estimating one demand parameter. Flexible demand systems, on the other hand, are difficult to work with due to an explosion in cross-price effects (dimensionality) and the challenges entry and exit pose for estimating elasticities (parameter stability). Placing mild restrictions on a translog demand system, I overcome the dimensionality and stability problems while retaining flexibility and tractability. I show that the translog entry/exit adjustment yields an aggregate average elasticity analagous to the CES case, but the underlying flexibility means this aggregate average may differ between entering and exiting goods and may change over time.Introducing Demographic Labor Market Data into the U.S. National Accounts
Abstract
The U.S. GDP and its foundational economic accounts contain some of the most widely used and followed economic statistics in the world yet contain limited information on the labor market and almost no information on demographic groups. We build a new demographic dataset using a small area estimation to deal with poorly estimated cells and integrate this data into the U.S. national accounts. The dataset includes labor market outcome data cross classified by sex, age, education, and industry. We show how this can be used to better understand relationships between economic growth and labor market outcomes by demographic group.Estimation of Race and Ethnicity by Re-Weighting Tax Data
Abstract
U.S. tax forms do not collect information about race or ethnicity. While no tax rules are based on the taxpayer’s race or ethnicity, not taking race and ethnicity into consideration in the policymaking process could result in the unintentional consequence of widening racial and ethnic disparities in after-tax income. In order to be able to evaluate this risk, we describe an imputation method developed to impute information about race and Hispanic origin (RH) to a stratified random sample of taxpayers. Specifically, we use a set of explanatory variables, including total income, filing status, age, number of dependents, taxable interest, presence of farm income, first name, last name, and the ZIP Code Tabulation Area of the residence, to make inferences about a taxpayer’s race and Hispanic origin. We apply Bayesian inference to estimate the probabilities that each taxpayer in our sample is in each of the 6 groups— Hispanic, White, Black, American Indian or Alaska Native, Asian or Pacific Islander, and multiple-race—given the variables, which, in turn, form the 6 RH weights for each taxpayer.U.S. Treasury Distribution Tables by (Imputed) Race
Abstract
U.S. tax forms do not collect information about race or ethnicity. While no tax rule is established based on the taxpayer’s race or ethnicity, not taking race and ethnicity into consideration in the policymaking process can result in the unintentional consequence of widening racial and ethnic disparities in after-tax income. Costello et al. (2021) imputed race to the Office of Tax Analysis’ Individual Tax Model by applying Bayesian inference to a set of explanatory variables available in tax data, including total income, filing status, age, number of dependents, taxable interest, presence of farm income, first name, last name, and the ZIP Code Tabulation Area (ZCTA). This paper extends their work by applying the imputed races and ethnicities to tables publicly available on the Department of the Treasury’s website. We find that White families are overrepresented in the top of the income distribution, relative to their share of the overall population, while Black and Hispanic families are over-represented in the bottom of the income distribution, relative to their share of the overall population. Disparities across groups within the same income decile in the amount of refundable credits received can be explained by differences in family structure and income sources.Removing Residual Seasonality from GDP
Abstract
We proposed three diagnostic tests in a recent study (JOS, 2022) for detecting stable and dynamic residual seasonality (RS) in the indirectly seasonal adjusted series. The three diagnostic tests were applied to quarterly real GDP and 19 aggregates from the U.S. National Income and Product Accounts (NIPA) using data from 1947 to 2018. The results revealed possible presence of RS in some of the NIPA aggregates in some time periods. In this study, we propose an optimization-based benchmarking solution method, extended from the framework developed by McElroy (2018), to remove RS from indirectly seasonal adjusted series via frequency- or cross-aggregation in the hierarchical tree structured national accounting system and reconcile the adjusted series according to the accounting constraints.Discussant(s)
John Haltiwanger
,
University of Maryland
JEL Classifications
- C8 - Data Collection and Data Estimation Methodology; Computer Programs