Estimating shadow economy in Tanzania: an analysis with the MIMIC approach

Abstract

Purpose The purpose of this paper is to estimate the Tanzanian shadow economy (SE) from 2003 to 2015 and test the statistical relationships between the SE and its potential causes and indicators. Design/methodology/approach The econometric analysis is based on a multiple indicators multiple causes (MIMIC) model. To calibrate the SE from the estimates, the authors adopt the value of 55.4 percentage of the SE to official GDP from the literature for the base year 2005. Findings The SE ranges from 52 to 61 per cent of official GDP and slightly decreases from 2013 to 2015. Increase in inflation, unemployment and government spending were the main drivers of the SE dynamics. Research limitations/implications Given the challenges facing estimation of the SE (e.g. small sample size, exogenous estimate to calibrate the model, meaning of the latent variable), quantification of SE should be considered to be rough measures. Practical implications To lower the size of the SE, the government needs to keep inflation and unemployment stable over time, to reduce government spending because it creates pressure on tax collection due to the limited tax base. Originality/value This is the first study specifically focused on Tanzanian SE based on the MIMIC approach. Existing estimates of Tanzanian SE are calculated by monetary models or apply a common MIMIC specification to the worldwide context.