Labor tax evasion is a major policy issue that is especially salient in transition and post-transition countries. In this brief, we use firm-level administrative data, tax authorities’ audit data and machine learning techniques to detect firms likely to be involved in labor tax evasion in Latvia. First, we show that this approach could complement tax authorities’ regular practices, increasing audit success rate by up to 35%. Second, we estimate that about 30% of firms operating in Latvia between 2013 and 2020 are likely to underreport the wage of (some of) their employees, with a slightly negative trend.
Tax evasion is a major policy issue that is especially salient in transition and post-transition countries. In particular, “envelop wage”, i.e., an unofficial part of the wage paid in cash, is a widespread phenomenon in Eastern Europe (European Commission, 2020). Putnins and Sauka (2021) estimate that the share of unreported wages in Latvia amounts to more than 20%. Fighting labor tax evasion is a key objective of tax authorities, which face two main challenges. The first is to make the best use of their resources. Audits are costly, so the choice of firms to audit is crucial. The second challenge is to track the evolution of the prevalence of labor tax evasion. For this purpose, most of the existing literature relies on survey data.
In our forthcoming paper (Gavoille and Zasova, 2022), we propose a novel methodology aiming at detecting tax-evading firms, using administrative firm-level data, tax authorities’ audit data and machine learning techniques.
This study provides two main contributions. First, this approach can help tax authorities to decide which firms to audit. Our results indicate that the audit success rate could increase by up to 20 percentage points, resulting in a 35% increase. Second, our methodology allows us to estimate the share of firms likely to be involved in labor tax evasion. To our knowledge, this paper is the first to provide such estimates, which are however of primary importance in guiding anti-tax evasion policy. We estimate that over the 2013-2020 period, about 30% of firms operating in Latvia are underreporting (at least some of) their workers’ wages.
The general idea of our approach is to train an algorithm to classify firms as either compliant or tax-evading based on observed firm characteristics. Tax evasion, like any financial manipulation, results in artifacts in the balance sheet. These artifacts may be invisible to the human eye, but machine learning algorithms can detect these systematic patterns. Such methods have been applied to corporate fraud detection (see for instance Cecchini et al. 2010, Ravisankar et al. 2011, West and Bhattacharya 2016).
The machine learning approach requires a subsample of firms for which we know the “true” firm behavior (i.e., tax-evading or compliant) in order to train the algorithm. For this purpose, we propose to use a dataset on tax audits provided by the Latvian State Revenue Service (SRS), which contains information about all personal income tax (PIT) and social security contributions (SSC) audits carried out by SRS during the period 2013-2020, including the outcome of the audit. The dataset also contains a set of firm characteristics and financial indicators, covering both audited and non-audited firms operating in Latvia (e.g., turnover, assets, profit). Assuming that auditors are highly likely to detect misconduct (e.g., wage underreporting) if present, audit outcomes provide information about a firm’s tax compliance. Firms sanctioned with a penalty for, say, personal income tax fraud are involved in tax evasion, whereas audited-but-not-sanctioned firms can be assumed compliant. The algorithm learns how to disentangle the two types of firms based on the information contained in their balance sheets. Practically, we randomly split the sample of audited firms into two parts, the training and the testing subsamples. In short, we use the former to train the algorithm, and then evaluate its performance on the latter, i.e., on data that has not been used during the training stage. If showing satisfying performance on the training sample, we can then apply it to the whole universe of firms and obtain an estimate of the share of tax-evading firms.
In this study, we successively implement four algorithms that differ in the way they learn from the data: (1) Random Forest, (2) Gradient Boosting, (3) Neural Networks, and (4) Logit (for a review of machine learning methods, see Athey and Imbens, 2019). These four data mining techniques have previously been used in the literature on corporate fraud detection (see Ravisankar et al. 2011 for a survey). Each of these four algorithms has specific strengths and weaknesses, motivating the implementation and comparison of several approaches.
Table 1 provides the out-of-sample performance of the four different algorithms. In other words, it shows how precise the algorithm is at classifying firms based on data that has not been included during the training stage. Accuracy is the percentage of firms correctly classified (i.e., the model prediction is consistent with the observed audit’s outcome). In our sample, about 44% of audited firms are required to pay extra personal income tax and social security contributions. This implies that a naive approach predicting all firms to be evading would be 44% accurate. Similarly, a classification predicting all firms to be tax compliant would be correct in 56% of the cases. This latter number can be used as a benchmark to evaluate the performance of the algorithms. ROC-AUC (standing for Area Under the Curve – Receiver Operating Characteristics) is another widespread classification performance measure. It provides a measure of separability, i.e., how well is the model able to distinguish between the two types. This measure is bounded between 0 and 1, the closer to 1 the better the performance. A score above 0.8 can be considered largely satisfying.
Table 1. Performance measures
Random Forest is the algorithm providing the best out-of-sample performance, with more than 75% of the observations in the testing set correctly classified. Random Forest is also the best performing model according to the ROC-AUC measure, with performance slightly better than Gradient Boosting.
Our results imply that a naive benchmark prediction is outperformed by almost 20 percentage points by Random Forest and Gradient Boosting in terms of accuracy. It is important to emphasize that this improvement in performance is achieved using a relatively limited set of firm-level observable characteristics that we obtained from SRS (which is limited compared to what SRS has access to), and that mainly come from firms’ balance sheets. This highlights the potential gain of using data-driven approaches for the selection of firms to audit in addition to the regular practices used by the fiscal authorities. It also suggests a promising path for further improvements, as in addition to this set of readily available information the SRS is likely to possess more detailed limited-access firm-level data.
Share of Tax-Evading Firms Over Time and Across NACE Sectors
We can now apply these algorithms to the whole universe of firms (i.e., to classify non-audited firms). Figure 1 shows the share of firms classified as tax-evading over the years 2014 to 2019 for our two preferred algorithms – Gradient Boosting and Random Forest. Random Forest (the best performing algorithm) predicts that 30-35% of firms are involved in tax evasion, Gradient Boosting predicts a slightly higher share (around 40%). Both algorithms, especially Random Forest, suggest a slight reduction in the share of tax-evading firms since 2014.
Figure 1. Share of tax-evading firms over time
The identified reduction, however, does not necessarily imply that the overall share of unreported wages has declined. In fact, existing survey-based evidence (Putnins and Sauka, 2021) indicate that the size of the shadow economy as a share of GDP remained roughly constant over the 2013-2019 period, and that there was no reduction in the contribution of the “envelope wages”. With our method, we are estimating the share of firms likely to be involved in labor tax evasion. Unlike the survey approach, our methodology does not allow the measurement of tax-evasion intensity. In other words, the share of non-tax compliant firms may have decreased, but the size of the envelope may have increased in firms involved in this scheme.
Next, we disaggregate the share of tax-evading firms by the NACE sector. Figure 2 displays the results obtained with Random Forest, our best performing algorithm.
Figure 2. Share of tax-evading firms by NACE, based on Random Forest
First, the sector where tax evasion is the most prevalent is the accommodation/food industry, where the predicted share of tax-evading firms is 70-80%. Second, our results indicate that the overall decrease in the share of firms likely to evade is not uniform. It is mostly driven by the accommodation/food and manufacturing sectors. Other sectors remain nearly flat. This highlights the fact that labor tax evasion varies both in levels and in changes across sectors.
We show that machine learning techniques can be successfully applied to administrative firm-level data to detect firms that are likely to be involved in (labor) tax evasion. Machine learning techniques can be used to improve the selection of firms to audit in order to maximize the probability to detect tax-evading firms, in addition to the regular practices already used by SRS. Our preferred algorithms – Random Forest and Gradient Boosting – outperform the naive benchmark classification by almost 20 percentage points, which is a substantial improvement. Once implemented, the use of these tools can improve the audit effectiveness at virtually no extra cost.
Our findings also suggest a promising path for further improvements in the application of such methods. The improvement in predictive power achieved by our proposed algorithm is attained by using a limited set of variables readily available from the firms’ balance sheets. Given that SRS is likely to have access to more detailed firm-level information that cannot be provided to third parties, there is clear room for improving the performance of the algorithms by using such limited-access data.
Acknowledgement: The authors gratefully acknowledge funding from the Latvian State Research Programme “Reducing the Shadow Economy to Ensure Sustainable Development of the Latvian State”, Project “Researching the Shadow Economy in Latvia (RE:SHADE)”; project No VPP-FM-2020/1-0005.
- Athey, Susan, and Guido Imbens. 2019. “Machine Learning Methods That Economists Should Know About.” Annual Review of Economics 11: 685–725.
- Cecchini, Mark, and Haldun Aytug, and Gary J. Koehler, and Praveen Pathak, 2010. “Detecting management fraud in public companies“. Management Science 56, 1146-1160.
- European Commission, 2020. “Undeclared Work in the European Union. Special Eurobarometer 498” (Report)
- Gavoille, Nicolas and Anna Zasova, 2022. “Estimating labor tax evasion using tax audits and machine learning”, SSE Riga/BICEPS Research papers, forthcoming.
- Putnins, Talis, and Arnis Sauka, 2021. “Shadow Economy Index for the Baltic Countries 2009–2020” (Report), SSE Riga
- Ravisankar, Pediredla, and Vadlamani Ravi, and Gundumalla Raghava Rao, and Indranil Bose, 2011. “Detection of financial statement fraud and feature selection using data mining techniques“. Decision Support Systems, 50(2), 491-500.
- West, Jarrod, and Maumita Bhattacharya, 2016. “Intelligent financial fraud detection: a comprehensive review“. Computers & security, 57, 47-66
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
It is well-documented that foreign-owned firms often pay higher wages than domestic firms. This phenomenon is usually explained by foreign firms being more productive. In this brief, we discuss another mechanism that drives the wage premium for employees of foreign-owned firms. By comparing income and expenditures of households led by employees of foreign-owned firms, domestic firms and public enterprises in Latvia, we show that employees of foreign-owned firms receive less undeclared cash payments than employees of domestic firms.
A vast economic literature documents a wage premium for employees of foreign-owned firms (e.g., Heyman et al., 2007; Hijzen et al., 2013). This can result from self-selection of foreign firms in highly productive sectors (Guadalupe et al., 2012) or from a productivity increase (Harding and Javorcik, 2012). In a recent paper (Gavoille and Zasova, 2021), we provide evidence of a third driver: foreign-owned firms are more (labor) tax compliant than domestic firms.
Envelope wage, i.e., an unreported cash-in-hand complement to the official wage, is a widespread phenomenon in transition and post-transition countries (e.g., Gorodnichenko et al., 2009 in Russia, Putninš and Sauka, 2015 in the Baltic States, Tonin, 2011 in Hungary). Employees are officially registered, but the income reported to tax authorities is only a fraction of the true income, the difference being paid in cash. If domestic firms are more likely to underreport wages than foreign-owned ones, the documented wage premium for employees of foreign-owned firms is overestimated.
Methodology and data
To compare the prevalence of income underreporting in foreign and domestic firms, we use an approach similar to Pissarides and Weber (1989). This approach is based on two main assumptions. First, even though households participating in an expenditure survey can have incentives to misreport their expenditures, they accurately report their expenditure on food.
The second assumption is that if all households would fully report their income, similar households would report a similar share of spending on food. If, however, a group of households is likely to underreport income, their fraction of income spent on food will systematically be higher than that of tax-compliant households. Using the propensity to food consumption of a group of households that cannot evade payroll tax as a benchmark, we can identify groups of tax-evading households by comparing their food consumption with the reference group.
In this brief, we mainly focus on three household groups: households where the head is an (1) employee of a foreign-owned firm (reference group), (2) employee of a public sector enterprise, and (3) employee of a domestic firm. We introduce public sector employees as an additional comparison group, since they cannot collude with employers to underreport wages. Hence, our approach allows us to test whether households in the third group are more likely to receive undeclared payment than households in the first group, and additionally test if our reference group is systematically different from public sector employees.
We estimate Engel curve-type relationships for food consumption for different types of households, i.e., we estimate how households’ food consumption varies with income depending on employment of the main breadwinner (employed in a foreign-owned firm, public sector enterprise, domestic firm or self-employed), controlling for various household characteristics (number of adults, size of household, place of residence, level of education of the main breadwinner, and other).
Our data comes from three sources. First, we use the 2020 round of the Latvian Household Budget Survey (HBS), which provides information on household consumption, income and characteristics in 2019. Second, we use an administrative matched employer-employee dataset providing information on reported wages for the whole population of employees in Latvia. We match the second database with HBS using (anonymized) individual IDs contained in both datasets. Finally, we use (anonymized) firm IDs contained in the second database to merge it with a third data source, which provides detailed information on firms’ foreign-ownership status.
For simplicity, in the rest of the brief we denote “household where the head is an employee of a foreign-owned firm” as simply “foreign-owned households”. A similar simplification applies to other household groups.
Comparing domestic and foreign-owned households, domestic households spend a higher share of their income on food. Figure 1 plots a non-parametric Engel curve for the two groups. The two curves exhibit fairly similar behavior, but the Engel curve for domestic households always lies above the one for foreign-owned households: for a given income, domestic households always spend a larger fraction on food than foreign-owned ones.
Our model estimations provide two main results. First, we find that the net wage premium for employees of foreign firms is 13-35%, depending on the sample and the source of data on income. Second, we show that domestic households are more likely to underreport income than foreign-owned households. On average, domestic firm households are estimated to conceal 26% more income than foreign-owned ones. At the same time, public sector households do not exhibit a significantly different food consumption pattern than foreign-owned firm households. Assuming that public sector households cannot evade, foreign-owned firm households hence do not underreport. The estimated share of concealed income is even larger (about 40%) if we restrict our sample to households where the head is aged below 50 years and is full-time employed.
Figure 1. Engel curve
In a context of widespread labor tax evasion, the observed wage premium for employees of foreign-owned firms can be driven by payroll tax compliance. How much of the wage premium can underreporting explain? Our results for Latvia suggest a net wage premium of 13% to 35% for the group of foreign-owned households. This roughly corresponds to the magnitude of the underreporting factor, indicating that nearly all of the wage premium can be explained by labor tax evasion. Even though the precise underreporting point estimates should be cautiously interpreted, and this 1-to-1 relation is anecdotal, this nevertheless highlights the potential importance of envelope wages in explaining the wage premium of employees of foreign-owned firms when labor tax evasion is prevalent.
Acknowledgement: This brief is based on a recent article published in Economics Letters (Gavoille and Zasova, 2021). The authors gratefully acknowledge funding from LZP FLPP research grant No.LZP-2018/2-0067 InTEL (Institutions and Tax Enforcement in Latvia).
- Gavoille, Nicolas; and Anna Zasova, 2021. “Foreign ownership and labor tax evasion: Evidence from Latvia”, Economics Letters, 207, 110030.
- Gorodnichenko, Yuriy; and Jorge Martinez‐Vazquez; and Klara Sabirianova Peter, 2009. “Myth and Reality of Flat Tax Reform: Micro Estimates of Tax Evasion Response and Welfare Effects in Russia“, Journal of Political Economy, 117 (3), pages 504-554.
- Guadalupe, Maria; and Olga Kuzmina; and Catherine Thomas, 2012. “Innovation and Foreign Ownership“, American Economic Review, 102 (7), pages 3594-3627.
- Harding, Torfinn; and Beata S. Javorcik, 2012. “Foreign Direct Investment and Export Upgrading“, The Review of Economics and Statistics, 94 (4), pages 964–980.
- Heyman, Fredrik; and Fredrik Sjöholm; and Patrik Gustavsson Tingvall, 2007. “Is there really a foreign ownership wage premium? Evidence from matched employer–employee data“, Journal of International Economics, 73 (2), pages 355-376.
- Hijzen, Alexander; and Pedro S. Martins; and Thorsten Schank; and Richard Upward, 2013. “Foreign-owned firms around the world: A comparative analysis of wages and employment at the micro-level“, European Economic Review, 60, pages 170-188.
- Hurst, Erik; and Geng Li; and Benjamin Pugsley, 2014. “Are Household Surveys Like Tax Forms? Evidence from Income Underreporting of the Self-Employed“, The Review of Economics and Statistics, 96 (1), pages 19–33.
- Pissarides, Christopher A.; and Guglielmo Weber, 1989. “An expenditure-based estimate of Britain’s black economy“, Journal of Public Economics, Volume 39 (1), pages 17-32
- Putninš, Tālis J.; and Arnis Sauka, 2015. “Measuring the shadow economy using company managers“, Journal of Comparative Economics, 43 (2), pages 471–490.
- Tonin, Mirco, 2011. “Minimum wage and tax evasion: Theory and evidence“, Journal of Public Economics, 95 (11–12), pages 1635-1651.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
This policy brief discusses residents’ voluntary payments for local public goods in Russian municipalities in a historic and a comparative international context. We emphasize the behavioral aspects of such collective action and the political economy risks of implementing this financial mechanism. Finally, we use data from the Russian Federal Treasury to offer empirical evidence on the regional variation in the amounts of these payments.
One of the drawbacks of a system of fiscal federalism is that it often results in an inadequate distribution of fiscal authority between regional and local governments. As a result, municipalities may be incapable of levying a due amount of taxes for the provision of the required quantity or quality of local public goods. A solution to the problem may be found in residents’ voluntary monetary and nonmonetary contributions to local projects. In Russia these are financial contributions to projects such as the renovation of roads, pedestrian bridges, parks, sports grounds and playgrounds, street lighting, the cleaning of ponds and rivers, etc. (Besstremyannaya, 2019). The experience of municipalities in the U.S. provides similar examples of residents’ financial contributions to municipal projects: here these resources are used, for example, as additional funds for financing secondary education (Winerip, 2003).
There are several potential theoretical explanations for the motivation of individuals to engage in such voluntary contributions. To start with, they can be motivated by purely altruistic concerns about local public welfare and the benefits of others (Ferris, 1984). Alternatively, the motives of individuals may be better (and, perhaps, more realistically) described by the approach of impure altruism (Andreoni, 1990), which trades off the amount of the public good against the size of contribution in motivating the behavior of the individual.
Further, the phenomenon of voluntary contributions is closely related to the desire of local authorities to substitute for insufficient budgetary revenues. As a result, residents may be coerced to submit monetary or in-kind labor payments for community development (Beard, 2007), for instance to put asphalt on rural roads (Olken and Singhal, 2011). Accordingly, instead of referring to residents’ contributions as donations, one may consider this mechanism of raising extra resources for local budgets as a type of informal tax. Arguably, altruism for the provision of public goods may be more prevalent among residents in developed countries, while contributions for local projects in developing countries reflect direct or indirect coercion by local authorities.
This policy brief analyses voluntary contributions to municipal budgets by residents in modern Russia. The presence of only fragmentary evidence from other countries limits our formal comparative analysis. However, we attempt to summarize common issues on the implementation of this financial mechanism.
Russian experience and legal framework
In contemporary Russia, residents’ contributions to local projects are called self-taxes (“samooblozheniye”) and constitute non-tax revenues of local governments (Article 131 of the Budget Code of 1998, Article 56 of the Local Government Law of 2003).
According to the Budget Code, self-taxes are fixed-size and one-time payments by residents for purposes of local projects, with the list of projects initiated directly by residents and voted at a local referendum. The projects commonly include activities on improving local infrastructure (Balynin, 2015). The December 2017 revision of the Local Government Law attempted to add additional incentives for self-financing by making it more targeted (in terms choosing urgent local projects) and easier to coordinate (in terms of organizing a referendum) by allowing referendums by residents of only selected parts of a municipality.
According to the Budget Code of the Russian Federation, self-taxes are earmarked items in non-tax revenues of local budgets, and may not be interrelated with other types of revenues or with the deficit of local government.
It should be noted that the use of self-taxes is not new in modern Russia: this funding mechanism was temporary exploited by Soviet authorities in the 1920s and 1930s, and was revitalized in the emerging Russian economy in the 1990s.
The early use of self-taxes in Soviet Russia illustrates the issue with the transformation of self-taxation into informal taxation. Self-taxation was introduced in 1924 as a formally voluntary decision of residents on financing local public goods. However, one may doubt whether the decision was indeed made by residents without pressure. Moreover, the lists of local public goods to be financed by self-taxes were determined by public authorities (Resolution by the Central Executive Committee and Soviet of National Commissars of the USSR of 3.08.1931). An illustration of the opposition to this informal tax may be found in the protocols of the council of residents of Roksanka, a village in Kaluzhskaya region from August 1928: citizens decided not to use self-taxes to finance a local school, since they believed there were sufficient budgetary funds – namely revenues from sold public property (Sergienko, 2015).
Calls to avoid similar retransformation of self-taxes into an informal tax were noted in modern Russia in 2006-2007, when the Bills on the amendments to the Local Government Law attempted to empower local authorities with the rights to deal with issues of self-taxation (Emetz and Makarov, 2016).
Private contributions in the form of monetary payments or labor participation are common in developing countries and are explained by the need to improve the insufficient quality of basic local public goods. For instance, the mechanism is used for road construction, water supply or primary education (Olken and Singhal, 2011). At the same time, residents of developed countries may choose to contribute to sustaining a high quality of local public goods: fire departments, medical centers, museums (Bice and Hoyt, 2000; Ferris, 1984). For example, the introduction of a redistributional mechanism of budgetary funds across rich and poor municipalities may lead to a decrease in the quality of public goods in the richest municipalities (owing to a fall in per capita funds after the redistribution). Accordingly, residents of rich municipalities may voluntarily decide to collect extra funds to recover the formerly high quality of public goods (e.g. secondary education in the U.S., see Brunner and Sonstelie, 2003; Winerip, 2003).
A common challenge to implementing a mechanism of voluntary payments is associated with the difficulties of reaching a decision within a large group of individuals. Indeed, residents may demonstrate selfish behavior or may follow selected local leaders (Jack and Recalde, 2015; Blackwell and McKee, 2003). Moreover, the common lack of enforcement instruments makes voluntary contributions unreliable (Slemrod, 1998).
Interestingly, the methods of dealing with non-compliance are similar across countries: the Perm krai of Russia, municipalities in the U.S. and villages in developing countries use techniques as such notification by mail, home visits, disclosing the lists of non-compliers and employing various ways of informal coercion by neighbors or public leaders (Olken and Sighal, 2011; Miguel and Gugerty, 2005; Winerip, 2003).
Data from Russian regions
We use the 2013-2016 annual data of the Russian Treasury, which allows disentangling self-taxes as an item in the list of non-tax revenues of local budgets. Owing to municipal-level data being unavailable, our analysis concerns the sum of local budgets in each region. Only 33 regions out of 83 analyzed regions had positive self-tax revenues in 2013, and the leading regions in 2015-2016 are the Tatarstan Republic, the Bashkortostan Republic, Kirovskaya oblast, Lipetskaya oblast, Kaluzhskaya oblast, and Perm Krai. The share of self-taxes in non-tax revenues is rather low: it amounts to 2-3% in Tatarstan, while it is less than 1% in the remaining regions (Table 1).
Table 1. Top regions according to levied self-taxes in 2015-2016
|Self-taxes in 2015||Self-taxes in 2016|
|Thousand rubles||% of non-tax revenues of local budgets||Thousand rubles||% of non-tax revenues of local budgets|
|Republic South Osetiya-Alaniya||1807.11||0.38||1451.16||0.30|
Arguably, the share of self-taxes in non-tax revenues is not associated with the desire to compensate for insufficient transfers from the federal or regional budgets: the absolute value of the correlation coefficient with the share of transfers to local budgets in non-tax revenues is below 0.25 (Besstremyannaya, 2019, Table 1). Similarly, we found no interrelation of the share of self-taxes with such socio-economic variables as (per capita) gross regional product and density of population.
Next, we focus on the policy of regional governments to provide budgetary funds on top of the money collected through self-taxes. As of 2016, such regional co-financing was present in the Tatarstan Republic, Kirovskaya oblast, Vladimirskaya oblast and Perm Krai (Emetz and Makarov, 2016; Balynin, 2015). The coefficient of regional co-financing of local projects (the amount of regional funds over the locally provided funds) equals 1 in Vladimirskaya oblast and varies from 1.5 to 5 in other above-mentioned regions. Examples of such co-provision of local public goods by region and municipalities include the renovation of water supply facilities in Perm Krai and the cleaning of lakes in Tatarstan (Nikitin, 2018, Platoshino budget, 2017).
Our estimates reveal that coefficients higher than 1 are associated with a higher prevalence of self-taxes. Indeed, the increase in the share of self-taxes in the revenues of local budgets in such regions is much higher than the corresponding growth in regions without co-financing or with unity co-financing (Besstremyannaya, 2019, Table 2).
Finally, we use the data for Perm Krai which experienced a recent reform with a rise of the coefficient from 3 to 5 in 2014. Our estimates of the treatment effect of such a reform and an extrapolation to other regions reveal that a unit increase of the coefficient causes a 55% growth in the share of self-taxes in non-tax revenues (Besstremyannaya 2019, Table 3).
To sum up, regional co-financing of local projects is associated with a growth in the collected self-taxes.
The phenomenon of voluntary contributions to local budgets is relatively common in real life. However, the literature addressing it is rather fragmented. In particular, little is known empirically on the motivation of individuals to engage in such contributions.
Our analysis with the 2013-2016 annual data for Russian regions reveals that residents’ contributions to local public goods are unrelated to insufficient revenues by local budgets. Moreover, the share of residents’ contributions in the budgetary non-tax revenues is positively associated with regional co-financing of these local projects. Hence, one may conjecture that in Russia, this phenomenon may be viewed as an altruistic attempt to raise quality of local public goods or as a means to signal about the most urgent local projects to regional governments.
- Andreoni, J. (1990). Impure altruism and donations to public goods: A theory of warm-glow giving. The Economic Journal, Vol. 100(401), pp. 464-477.
- Balynin I.V. (2015). The use of self–taxes of citizens in forming the revenues of the local budgets. Finansy i Upravleniye, No.2, pp. 53–62. (In Russian).
- Beard, V. A. (2007). Household contributions to community development in Indonesia. World Development, Vol. 35(4), pp. 607-625.
- Besstremyannaya G.E. (2019) Informal taxes for the provision of public goods in Russian regions. Voprosy Ekonomiki No.1, pp 124-134. (In Russian).
- Bice D.C., Hoyt W.H. (2000). The impact of mandates and tax limits on voluntary contributions to local public services: An application to fire–protection services. National Tax Journal, Vol. 53(1), pp. 79–104.
- Blackwell C., McKee M. (2003). Only for my own neighborhood?: Preferences and voluntary provision of local and global public goods. Journal of Economic Behavior and Organization, Vol. 52(1), pp. 115–131.
- Brunner E., Sonstelie J. (2003). School finance reform and voluntary fiscal federalism. Journal of Public Economics, Vol. 87(9–10), pp. 2157–2185.
- CEFIR project for the Ministry of Finance of the Russian Federation “The development of methodological recommendations for increasing the revenues of Russian regions and municipalities” (Final report of November 2017)
- Emetz M.I., Makarov M.A. The self–taxation of citizens as a source of local budget revenues. Ekonomika i Menedgment Innovatsionnyh Tehnologii, No. 12, http://ekonomika.snauka.ru/2016/12/13433 (In Russian).
- Ferris J.M. (1984). Coprovision: Citizen time and money donations in public service provision. Public Administration Review, pp. 324–333.
- Jack B.K., Recalde M.P. (2015). Leadership and the voluntary provision of public goods: Field evidence from Bolivia. Journal of Public Economics, Vol. 122, pp. 80–93.
- Miguel E., Gugerty M.K. (2005). Ethnic diversity, social sanctions, and public goods in Kenya. Journal of Public Economics, Vol. 89(11–12), pp. 2325–2368.
- Nikitin, E. (2018) Self-taxation in Tatarstan republic provides for 4 to 1 budgetary cofinancing https://www.tatar-inform.ru/news/2018/02/16/593910/
- Olken B.A., Singhal M. (2011). Informal taxation. American Economic Journal: Applied Economics, Vol. 3(4), pp. 1–28.
- Platoshino budget for 2017 and forecast for 2018-2019. http://www.platoshino59.ru/index.php/2012-01-05-09-21-25/2016-08-28-09-18-38/item/403-publichnyj-byudzhet-platoshinskogo-selskogo-poseleniya-na-2017-god-i-planovyj-period-2018-i-2019-godov-byudzhet-dlya-grazhdan
- Sergienko, N.S. (2015). Self-taxation as a form of voluntary participation of the population in socioeconomic development of settlements. Sovremennye Issledovaniya Sotsialnyh Problem, Vol. 1, No. 21, pp. 266–270. (In Russian).
- Slemrod, J. (1998). On voluntary compliance, voluntary taxes, and social capital. National Tax Journal, Vol. 51(3), pp. 485–491.
- Winerip, M. 2003. On Education: Giving green or turning red. The New York Times, Feb 26.
Belarus proudly calls itself a social state. Indeed, Belarus boasts one of the lowest poverty and inequality levels in the region. Fiscal policy in Belarus is equalizing and pro-poor, effectively redistributing income from rich to poor. As in Russia and many other Post-Soviet states, the equalizing effect of the fiscal policy in Belarus is mostly attributable to the pension system. Some of the other social policies are highly inefficient, failing to redistribute income. The prominent examples are utility subsidies and student stipends, which mainly benefit the upper part of the income distribution. The lack of adequate unemployment benefits is an opportunity to improve the efficiency of the social support system in Belarus.
The Constitution of Belarus characterizes Belarus as a social state, and Belarus takes its social state status seriously. The economic growth in the beginning of the 2000’s was strongly pro-poor (Chubrik, 2007). Poverty according to the national definition (calorie-based poverty line, which in 2015 corresponded to $10.67 PPP per day) declined from 42% in 2000 to 5.7% in 2016, while the poverty according to the international threshold of $3.1 per day in PPP terms is fully eradicated. Belarus also has one of the lowest levels of income inequality in the region with a Gini coefficient of only 0.27 (UNDP, 2016).
How much of the pro-poor and equalizing effects could be attributed to the government policy? Probably it is impossible to give a complete answer to the question. Many non-formalized and not easily quantifiable government policies lead to the decrease in poverty and inequality. For example, the policy of support to state-owned enterprises might have redistributive effects through job creation. However, the absence of access to relevant data makes it impossible to estimate the effects of the policy.
Some of the government policies, on the other hand, are easily quantifiable with available data. Bornukova, Chubrik and Shymanovich (2017) analyze the redistributive effects of fiscal policies in Belarus using the Commitment to Equity methodology (Lustig, 2016). The authors find that the direct taxes and transfers in Belarus (taxes, transfers, and subsidies) are equalizing and pro-poor, lowering the national poverty headcount by 17 percentage points and the income Gini coefficient from 0.41 to 0.27. The high equalizing effect of the fiscal policies in Belarus surpasses those in other developing countries, including Russia where the direct taxes and subsidies reduced the income Gini coefficient by 0.13 (Lopez-Calva et al., 2017). The remaining discussion in this brief is based on the results from Bornukova, Chubrik and Shymanovich (2017), if not otherwise stated.
Fiscal policies and their redistributive effects
The two types of direct personal taxes – the personal income tax and the social contributions tax – are both almost flat in Belarus. To fight tax evasion, the Belarusian authorities introduced flat tax rates in 2009, following a successful experiment in Russia. The personal income tax has some small exemptions for families with children, while the social contributions tax has a lower rate for agriculture employees. However, the effect of these deductions is relatively small: the direct taxes decrease the Gini coefficient by only 0.015.
The indirect taxes – the value-added tax, the import duties, and the excises – are weakly regressive, putting the burden of taxation on the poor. This is particularly true for the alcohol and tobacco excises. Again, the main purpose of these taxes is to penalize unwelcome behavior, and not to redistribute income, hence the result is not unexpected, and common for many countries. Overall the indirect taxes in Belarus increase the Gini coefficient by 0.05.
Direct transfers are responsible for most of the equalizing effects of the fiscal policies. This is not surprising, given that the main purpose of the direct transfers is to fight poverty and provide support for those in need. However, most of the transfers are not need-based or targeted to the poor. Instead they are assigned to households based on their socio-economic characteristics aside income, such as age and maternity status.
Pensions are the main factor of reducing poverty and inequality. They reduced the Gini coefficient by 0.11 and decreased poverty (according to national definition) by 19 percentage points. The incredible effectiveness of the pensions is largely explained by the absence of other sources of income of the retirees. The majority of them does not work, and have no other pension savings or passive income. Pensions in Belarus are also redistributive in nature since they only weakly depend on one’s income during the working life.
Different benefits and privileges also decrease poverty and inequality, but at a much smaller scale. The childcare benefits (for families with children aged 0-3 years) contribute most to the effects, decreasing the Gini coefficient by 0.013 and poverty by 3 percentage points. The variety of privileges does not contribute much due to their relatively small size.
Utilities and transport subsidies are also important elements of the social support system, and their existence is usually justified by the necessity to support those in need. Since the utilities subsidies are incorporated into tariffs and available for everyone independent of need, they are in fact benefitting the rich (i.e. people with big apartments and houses).
Figure 1. Incidence of utilities subsidies by income deciles
As seen on Figure 1, upper deciles receive more support through utilities subsidies, and this support is quite substantial, often surpassing $1 per day in PPP. However, as a share of income the utilities subsidies are still progressive, and they in fact decrease the Gini coefficient by the tiny amount of 0.006, and decrease poverty (as any handout). The same is true for transport subsidies.
What could be improved?
Due to the flat nature of direct taxation and an absence of well-targeted needs-based transfers, some of the people in need still fall through the cracks. 1.9% of the population actually becomes poor after we account for the direct taxes and transfers. This headcount increases to 3.3% if we account for indirect taxes.
Another important issue is the efficiency of government transfers and subsidies in fighting poverty and inequality. It is not surprising that pensions have the largest equalizing contribution, as the government spends almost 11% of GDP on pensions. If we account for this fact and look at the efficiency (effect on poverty and inequality per dollar spent), pensions are not the leading program. It is in fact surpassed by different kinds of child support. Given that mothers in Belarus are allowed to take 3 years of unpaid maternity leave, which decreases household income, childcare benefits are relatively efficient.
The unexpected leader in efficiency is unemployment benefits, despite (or maybe due to) their negligible size. Shymanovich (2017) shows that unemployed face high risks of poverty, suggesting that an increase in the size of unemployment benefits and an easier access may bring huge benefits. The current minuscule size of the benefits (around $10-15 per month) is still enough to lift some people out of poverty, and has important equalizing effects, generating the biggest “bang for the buck” out of all benefits.
The student grants (stipends), the utilities subsidy and the transport subsidy have very low efficiency. These programs relocate a lot of funds to the upper deciles of the income distribution. Our calculations show that if all benefits, privileges and subsidies were not available to those in the top two income deciles, the Belarusian budget could save 1.4% of GDP.
Fiscal policies in Belarus are quite effective in redistributing income. Bornukova, Chubrik and Shymanovich (2017) show that the direct taxes and transfers in Belarus result in a decrease of poverty by 17 percentage points, and decrease the Gini coefficient of inequality from 0.41 to 0.27. The pension system has the most important contribution, decreasing poverty by 19 percentage points, and the Gini coefficient by 0.11.
However, the absence of a needs-based, well-targeted social support system leads to many inefficiencies. Direct and indirect taxes lead to impoverishment of 3.3% of population, which is not compensated by direct transfers.
The absence of targeting also leads to 1.4% of GDP redistributed towards the two upper income deciles through benefits, privileges and subsidies. This is, of course, highly inefficient. Better targeting could allow saving these funds or redirecting them to unemployment benefits – the most efficient but a very small benefits program so far.
- Bornukova, Kateryna, Alexander Chubrik and Gleb Shymanovich, 2017. “Fiscal Incidence in Belarus: a Commitment to Equity Analysis”, BEROC Working Paper Series, WP no. 42
- Chubrik, Alexander, 2007. “GDP Growth and Income Dynamics: Who Reaps the Benefits of Economic Growth in Belarus?” In Haiduk, K., Pelipas, I., Chubrik, A. (Eds.) Growth for All? Economy of Belarus: The Challenges Ahead; IPM research Center
- Lopez-Calva, L. F., Lustig, N., Matytsin, M., Popova, D., 2017. “Who Benefits from Fiscal Redistribution in Russia?”,in The Distributional Impact of Fiscal Policy: Experience from Developing Countries, edited by Gabriela Inchauste and Nora Lustig (Washington: World Bank, forthcoming).
- Gleb Shymanovich, 2017. “Poverty and Vulnerable Groups in Belarus: The Consequences of 2015-2016 Recession (in Russian)”, IPM Research Center Bulletin
- UNDP, 2016. “Regional Human Development Report 2016: Progress at Risk”, United Nations Development Programme, Istanbul Regional Hub, Regional Bureau for Europe and the CIS