Data and Model Cross-Validation to Improve Accuracy of Microsimulation Results: Estimates for the Polish Household Budget Survey

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in AbstractWe conduct detailed analysis of the Polish Household Budget Survey data for the years 2006-2011 with the focus on its representativeness from the point of view of microsimulation analysis. We find important discrepancies between the data weighted with baseline grossing-up weights and official statistics from other sources. A number of re-weighting exercises is examined from the point of view of the accuracy of microsimulation results and we show that using a combination of demographic calibration targets with several economic status variables or tax identifiers from the microsimulation model substantially improves the correspondence of model results and administrative data. While demographic re-weighting is neutral from the point of view of income distribution, calibrating the grossing-up weights to adjust for economic status and tax identifiers significantly increases income inequality. We argue that although data reweighting can substantially improve the accuracy of microsimulation it should be used with caution.

Reevaluating Distributional Consequences of the Transition to Market Economy in Poland: New Results From Combined Household Survey and Tax Return Data

“…Similar discrepancies were found in case of subpopulations defined with respect to the education level, employment type or type of household residential area. Following Creedy (2004) and Deville and Särndal (1992), Myck and Najsztub (2015) propose to address this problem by calibrating the PHBS baseline weights using information from various administrative sources. This reweighting approach leads to adjusted weights that allow for obtaining weighted PHBS estimates corresponding closely to values taken from administrative sources (such as the official number of children in the population or people with higher education).…”

“…In this paper, we follow Myck and Najsztub (2015) in using the reweighing approach to adjust the baseline PHBS weights. In particular, we use two types of weight calibration.…”

“…For this period, Bukowski and Novokmet (2017) estimate top income shares using tabulations on the settlement of the PIT published annually by the Ministry of Finance.10 The tabulations contain information on the number of taxpayers, the amount of income, and tax paid by income brackets as defined by the tax rate schedule. This information is highly grouped as limited progressivity of the Polish income tax system implies that the number of income brackets in the tax rate schedule is small (three for 1994- 7 Myck and Najsztub (2015) show that calibrating the baseline PHBS weights with respect to the total number of PIT payers significantly increases the level of income inequality in Poland over 2006-2011 as measured by the Gini index and other indices. 8 We use the modified OECD equivalence scale.…”

“…Second, we make a methodological contribution to the literature by showing that two techniques of adjusting income distributions (Pareto imputation of top incomes and survey reweighting) that have been so far applied separately can be implemented jointly to produce top-corrected inequality estimates that provide best fit to the data. To this end, we integrate the methodology of imputing to household survey data top incomes from Pareto models fitted on tax data (Bartels and Metzing 2018) with microsimulation-based survey reweighting approach using tax and other administrative information (Myck and Najsztub 2015).…”

Gender Pay Gap in Poland

“…Secondly, some authors argue that the methodology of the Polish Household Budget Survey (HBS) -a survey that supplies the EU-SILC dataset -can lead to bias. Myck and Najsztub (2015) claim that alternative methods of calibration of the HBS could be used and that they change the estimates of inequality measures significantly. The authors conducted analysis of representativeness of the HBS data for the years 2006-2011 and conclude that the weighing method used for calibration influences the estimates of income inequality, and the use of alternative methods leads to values of inequality significantly higher than those calculated on the basis of an existing dataset.…”