The quality of tax administration and firm performance: evidence from developing countries

Similarly, post-filing procedures (e.g., claiming a VAT refund, undergoing a tax audit, or appealing a tax assessment) can be more challenging for small taxpayers and younger and less-experienced firms. In addition, larger firms can also benefit from economies of scale due to specialization within firms.

Hsieh and Klenow (2014) find that 35-year-old plants were on average nine times more productive in the manufacturing sector in the USA. Evidence from developing countries also suggests that new firms generally exhibit lower productivity growth than incumbents (Li and Rama 2015).

The construction of the index on the strength of tax administration is similar to indices of other fiscal institutions and processes including budget institutions (Dabla-Norris et al. 2010) and public investment efficiency (Dabla-Norris et al. 2011).

See http://​www.​tadat.​org/​ for details. TADAT assessments are conducted at the request of a country’s Ministry of Finance or tax authority and focus on the administration of the major direct and indirect taxes that are critical to central/federal government revenues. The assessments are evidence based, with a team of TADAT assessors conducting interviews with the authorities, visiting tax offices in the headquarters and field, and examining documents, processes, and IT systems. The assessments are typically conducted by a team of four certified TADAT assessors. The latter are persons with at least 5 years of professional experience in tax policy or administration and who have pursued the online TADAT training and passed the online TADAT exam. The TADAT course ensures that the assessors have a thorough understanding of the TADAT methodology. The assessment is conducted over a period of 2 weeks. A written report is prepared and submitted to the authorities at the end of the assessment.

See http://​data.​rafit.​org for details.

Using cost of collection indicators that express tax revenue in terms of the overall cost of tax administration is in principle subject to the same criticism.

Another strand of the literature examines optimal tax enforcement (see Keen and Slemrod 2017; Creedy 2016). Related to this, several empirical papers examine the effects of particular types of tax administrative intervention related to enforcement on tax compliance (see Brockmeyer et al. 2016, for a brief survey). Other studies have investigated the relationship between tax capacity and subsequent economic growth and development (Gaspar et al. 2016).

Similar relationships obtain between compliance costs and other measures of size, such as assets or employment.

E-filing is associated with higher tax compliance costs if it is mandatory and not optional, if there is paper-based reporting together with e-filing, and if processes are complex (Eichfelder and Vaillancourt 2014).

Under the assumption of expected utility maximization (Andreoni et al. 1998), taxpayers will not pay tax as long as the cost of compliance exceeds the net benefit of noncompliance (see also Slemrod and Yitzhaki 2002).

A risk-based approach, for instance, takes into consideration different aspects of a business, such as historical compliance, industry- and firm-specific characteristics, and the size of a business, in order to better assess which businesses are most prone to tax evasion (IMF 2015). From a game-theoretic perspective, a risk-scoring mechanism using all the information provided by the taxpayer as well as their profile makes it more difficult for taxpayers to consistently underreport income and avoid audit (Alm and McKee 2004).

For instance, the use of electronic tracing of payments by the National Tax Service of Korea promoted the use of electronic payments and credit cards. The positive gains in terms of voluntary compliance and GDP growth have been significant (Sung et al. 2017).

Tax ratios and survey measures of willingness to comply are negatively correlated with corruption (OECD 2013).

For example, where a tax administration is unable to produce basic numerical data for purposes of assessing operational performance (e.g., in areas of filing, payment, and refund processing).

PCA transforms correlated variables into a smaller number of uncorrelated variables called principal components ranked according to how much variability they capture in the data. We use the first principal component, which is the one with the most variability.

We chose the 33 measurement dimensions that directly affect interactions between the tax administration and the taxpayers and could plausibly affect firm performance. The other 14 dimensions do not plausibly affect tax administration–taxpayer interactions. For example, how the tax administration collaborates with the Ministry of Finance in forecasting revenue estimates is an important measure of the tax administration's performance, but does not affect its relationship with taxpayers.

Levinsohn and Petrin (2003) offer a semi-parametric estimation technique that uses intermediate inputs used by firms as a proxy for unobserved productivity shocks.

Additional results based on the TAQI2 which is constructed using a different type of weighting, namely the simple average of all 33 dimensions, are qualitatively similar and available upon request.

Multicollinearity considerations prevent us from including interactions with all the subindices in the full regression model.

The countries include Argentina, Barbados, Belize, Bolivia, Bosnia and Herzegovina, Brazil, Bulgaria, Chile, Colombia, Costa Rica, Côte d'Ivoire, Dominican Republic, Ecuador, Ethiopia, FYR Macedonia, Ghana, Guatemala, Honduras, Jordan, Kosovo, Madagascar, Mauritius, Mexico, Nicaragua, Paraguay, Peru, Senegal, Tanzania, Uganda, and Uruguay.

The fact that these results are qualitatively similar to the ones using our overall index of tax administration could also reflect the fact that the electronic filing rate may very well be a proxy of other initiatives of the tax administration that reduce tax compliance burdens—for example, a ‘client focus’ of the tax administration, well-established taxpayer services, and in some cases the provision of pre-populated tax return forms.