Formal and informal institutions: understanding the shadow economy in transition countries

Abstract

This paper reviews work that tests (1) how formal and informal institutions, and especially their interaction, affect participation in the shadow economy in transition countries; and (2) how participating in these shadow economies affects individuals' well-being. The key findings are that a clash of individuals' perceptions of formal institutions with their informal institutions increases involvement in the shadow economy. Conversely, a trustworthy relationship with the government and other individuals makes people more inclined to comply. The importance of their social and institutional context also appears in how individuals' involvement in the shadow economy relates to their well-being. These findings complement insights from the rich literature on tax morale, on the exchange between public institutions and citizens and between culture and institutions more generally. The findings also contribute to the institutional economic literature by empirically showing that: (1) focusing on formal institutions alone, that is strengthening the rule of law, is a necessary but insufficient response to the shadow economy; (2) taking informal institutions, such as individuals' trust and tax morale, into account is of equal importance; and (3) most importantly, formal and informal institutions go hand in hand, and their interaction should be an essential part of the new institutional perspective.

1. Introduction

All countries experience some degree of the shadow economy, also known as the informal sector, underground economy or grey economy.¹ Yet, some countries have a larger shadow economy than others (Kelmanson et al., 2019; Medina and Schneider, 2018). In the Western world, transition countries (i.e. former communist countries) are amongst those with the largest shadow economies.² During the final years of communism and after its fall, these countries experienced a slowdown of growth that later led to growth contraction, a massive output fall and rising unemployment (Campos and Coricelli, 2002). They also witnessed a complexity of institutional and societal changes during transformation. As a consequence, their shadow economies increased significantly (Eilat and Zinnes, 2002; Schneider and Enste, 2000), which in turn can lead to negative implications reaching all levels of society. At the macro level, these implications include reduced tax revenues, budget deficits and a lower quantity and quality of public goods. At the meso level, think of competition between compliant and non-compliant firms and organizations; while at the micro level, individuals can experience lower pensions, reduced social security benefits in the case of losing a job, limited entitlement to loans and credits and an overall lower well-being. Ultimately, shadow economies can affect the overall trust in state authority and in others at the societal level.

In this paper, we focus on the micro level and try to offer an overview of the empirical research on drivers of informality at the individual level in transition countries. The thread underlying the overviewed research is their new institutional approach, where institutions are seen as systems of established and prevalent social rules that structure social interactions (Hodgson, 2006). Importantly, they employ the distinction between formal institutions such as laws related to labour arrangements or taxes, on the one hand (Calcagno and Lopez, 2017), and on the other, informal institutions that are those that fall outside the scope of laws, e.g. norms or tax morale (Andriani et al., 2022).

This distinction involves a theoretical approach that provides an important framework for studying the shadow economy in former communist countries, because institutional change is the defining feature of transition economies: ‘any inquiry into the causes and ultimate consequences of the transition must include an analysis of the incentives and sanctions governing various types of underground economic activities’ (Feige, 1997: 24). However, in the late 1990s, there was little empirical evidence on the relationship between institutions and individuals' involvement in the shadow economy, and hardly any in transition countries. The research we present below includes some of the pioneering work that empirically tests North's (1990) and Feige's (1997) theoretical contributions by recognizing not only the importance of formal and informal institutions for understanding the shadow economy, but by also considering their interaction and enforcement. In a nutshell, the main argument of this new institutional perspective is that a proper functioning of the economy needs the formal and informal institutions to be complementary and well enforced. When they are not, transaction costs increase, economic performance decreases and non-compliant behaviour prospers (Feige, 1997; North, 1994; Pejovich, 1999). Such outcomes are more likely to be expected in transition countries due to the incompatibility between the radically changed formal institutions and the more resistant informal institutions (Helmke and Levitsky, 2003; Pejovich, 1999). Against this theoretical backdrop, the main contribution of the empirical research we discuss here is in showing: (1) whether and how both types of institutions, their interaction and their enforcement affect individuals' participation in the shadow economy in transition countries and (2) how, in turn, participating in these shadow economies affects individuals' well-being.

Most of the empirical research presented below is based on Albania as a case study. Albania offers an ideal environment to test the new institutional perspective. Contrary to other transition countries, which had carefully introduced some form of free-market institutions during communism, Albania had remained completely isolated and was the last country to allow democratic changes at the beginning of the 1990s. In other words, at the onset of the transition period, the contrast between formal and informal institutions in Albania was rather black and white. The change from former communist institutions to the new ‘private-property-based’ institutions was radical, while the historically ingrained informal institutions were still in place. At the same time, the country fell into a deep economic crisis, which was accompanied by a rapid increase of the shadow economy (Gërxhani, 2009). For all these reasons, Albania offers an optimal ‘test bed’ of whether participation in the shadow economy is related to the incompatibility between formal and informal institutions. To contrast the empirical findings from an ‘ideal context’ such as in Albania, we also discuss some empirical studies from other transition countries that were not as extreme in their economic and institutional background (Beck and Laeven, 2006; Tridico, 2013).

Finally, the overview focuses on two aspects of the shadow economy, namely tax evasion and informal work. Tax evasion refers to ‘illegal actions to reduce tax liability’ (Slemrod, 2019: 906), while informal work is defined as ‘all jobs in informal sector enterprises, or all persons who, during a given reference period, were employed in at least one informal sector enterprise’ (Hussmanns, 2005:2). Various studies have shown that evading taxes is an important motive for individuals engaging in informal work, especially in transition countries (Franic, 2020; Gërxhani, 2004a; Meriküll and Staehr, 2010).

The rest of this paper is structured as follows: in the next section we discuss the individual and structural determinants of participation in the shadow economy. Such determinants include the interaction between formal and informal institutions, opportunities to evade taxes, and differences across gender, education and trust. We present the effects of participating in the shadow economy on individuals' well-being in section 3, followed by the concluding section.

2. Individual and structural determinants of participating in the shadow economy

2.1 Institutional interaction and the shadow economy

After the breakdown of communism, many former communist countries entered a transition phase from a planned to a free-market economy. This transition phase was characterized by a fall in output, growth contraction, high rates of unemployment, poverty and social insecurity, which were rapidly being buffered by an increasing informal sector. At the same time, the economic transition was accompanied by a complete overhaul of the former formal governing institutions. This led to a paradox of formal institutions changing rapidly and radically, while informal institutions – the norms and traditions established during 45 years of communism – changed much less so. As argued by North (1994: 366), norms provide ‘legitimacy’ to formal rules and a revolutionary change in formal institutions will be hampered by norms that lag behind. In line with North's argument, Feige (1997) posited the hypothesis that a clash between formal and informal institutions would lead to more non-compliant behaviour and thus more underground activities.

The study by Gërxhani (2004b) was probably the first study to use micro-economic data to test this hypothesis. In 2000, she conducted a field survey in an urban area of Tirana, the capital of Albania, which provided an ideal environment to empirically test Feige's hypothesis and contribute to the new institutional approach (see Gërxhani, 2004b for a detailed description of the study). The method used was a ‘self-administered’ questionnaire, also known as a ‘hand-delivered and collected questionnaire’ (Tindigarukayo, 2001). This method combines a structured questionnaire with the personal touch of: (1) explaining face-to-face the main aspects of the questionnaire to the participants, and (2) personally collecting the completed questionnaire at a later agreed-upon time. A total of 1,500 households were randomly selected to participate, with the main income earner being targeted as the respondent. Of this sample, 89.3% completed the questionnaire.³ For a detailed description of the field survey and the questionnaire, see Gërxhani (2007a).⁴

The data contain information on individuals' monthly income and taxes, and this information was used to estimate the dependent variable – individuals' extent of tax evasion with respect to personal and business income. The data also contain information on individuals' attitudes towards a series of statements related to formal and informal institutions. These attitudes were the basis of the empirical measures of people's perceptions of formal institutions and people's tax-related social norms (i.e. tax morality). These measures were included in the empirical analysis in their own right, but were also used to create the main independent variable, which was a measure of the ‘institutional clash’ between people's perceptions of formal and informal institutions. Finally, the questionnaire collected information on individuals' socio-economic and demographic attributes, such as age, education, gender and sector of work, which were used as controls.

The results of the main regression used to test Feige's hypothesis are summarized in Table 1.

Table 1. Institutions and tax evasion

Notes: The table summarizes the results of an ordered Probit regression where the dependent variable is a measure of the extent of evasion of the income tax concerned. It is measured on a discrete five-point scale (cf. Gërxhani, 2004b). ‘No’ indicates that there is no statistically significant effect of the variable concerned. ‘Yes’ indicates a positive and statistically significant effect.

A key finding of this study is that the variable that measures the clash between the two types of institutions has a positive and statistically significant effect on the evasion of both income tax types.⁵ This means that more tax evasion behaviour is observed when the perception of formal institutions clashes with individuals' informal institutions. This finding supports Feige's hypothesis. Even when formal or informal institutions themselves are not (both) related to tax evasion, the interaction between the two is related to tax evasion. This is precisely what a later study (Williams et al., 2014) found in Bulgaria. Based on Eurobarometer data from 1,018 face-to-face interviews conducted during 2013, they report that the greater the non-alignment of informal institutions with formal institutions, the higher the participation in the Bulgarian shadow economy. The consistent findings in these two studies confirm the importance of simultaneously considering both types of institutions when studying individual engagement in the shadow economy. Doing so is most pertinent for transition countries that have undergone a radical institutional restructuring.

2.2 Gender and the shadow economy

Does the relationship between individuals' perceptions of formal and informal institutions and their participation in the shadow economy vary according to individuals' attributes, such as gender? Empirical research shows that women evade taxes less often than men do, irrespective of whether one looks at developed countries (e.g. Doerrenberger and Peichl, 2022; Flathmann and Sheffrin, 2003; Franic and Cichocki, 2022; Giese and Hoffman, 1999; Kedir et al., 2011; Torgler, 2002) or transition countries (Hanousek and Palda, 2002; Merikull and Staehr, 2010). While some studies in the mid-2000s hinted at gender differences in tax morale⁶ (e.g. Molero and Pujol, 2004), little was known about the reasons underlying gender differences in tax evasion.

Using the data set described above, Gërxhani (2007b) tries to find out more by asking two research questions: (1) are there gender differences in tax evasion behaviour in Albania?, and (2) if yes, is there an institutional explanation? The second question is motivated by the new institutional approach and implies that gender differences in tax evasion behaviour could be explained by the different ways in which men and women participating in the informal sector experience formal and informal institutions and their interaction.

The results of this study show that women evade both personal and business income tax less frequently than men do, and the differences are statistically significant at the 5% level. Turning the attention to the role of institutions, women and men do not appear to differ much in their experience of a clash between formal and informal institutions. This is also reflected in how institutions affect their tax evasion behaviour; at the 5% level, women do not react differently to men to formal or informal institutions, nor to the clash between the two. Therefore, Gërxhani (2007b) concludes that the new institutional theory does not seem to explain much of the gender difference in tax evasion in Albania. Instead, the segregation of the labour market in Tirana may provide a better explanation. Female main income earners are largely concentrated in the state sector, which leaves little room for evasion (taxes are automatically withdrawn from one's monthly salary). Male main income earners, on the other hand, work mainly in the unofficial private sector and in their own small businesses, two parts of the economy that are highly associated with higher tax evasion. However, the author rightly questions whether women are intrinsically less inclined to evade taxes and therefore self-select into public sector jobs, or whether the sex segregation of the labour market is the result of historical, economic or institutional reasons, which ultimately affect both men and women's evasion behaviour.

Cichocki and Tyrowicz (2010) report on gender differences as part of their investigation into the individual drivers of informal employment in Poland. They use a compilation of two data sets, based on a survey on unregistered employment, conducted by the Centre for Social and Economic Research for the Polish Ministry of Labour and Social Affairs in 2007. The results show that women evade taxes less frequently than men, by engaging less in informal work. When they do engage in informal work, the wage gender differences are not significant. Although the study does not provide an explanation for why women are less involved in informal activities, they do conclude that the engagement in informal work – which is higher among Polish men – is mainly driven by a lack of alternative employment opportunities.

2.3 Education and the shadow economy

Is participation in the shadow economy related to one's education level? Research in transition countries argues that this is likely to be the case. The direction of this effect, however, is under debate. While some say that a higher education level may increase participation because of a lack of alternative employment possibilities and the corresponding financial needs (e.g. Dimova et al., 2005), others emphasize the importance of education in developing civic norms and engagement with society (e.g. Nunn et al., 1978), which in turn may discourage people to participate in non-compliant activities.

Based on the same observational data set as in Gërxhani (2004b), collected in the urban area of Tirana in 2000, Gërxhani and van de Werfhorst (2013) empirically examine: (1) the relationship between individuals' education – while considering other individual, social and institutional factors – and their participation in the shadow economy, and (2) the channels – income-related or civic-norms-related returns of education – through which this relationship is established. The authors create their dependent variable on informal sector participation⁷ by combining information collected about respondents' tax evasion with information about the sector in which they work (i.e. distinguishing between jobs with and without a labour contract). Individuals' education level is the independent variable. After examining whether these two variables are related, the authors test whether the relationship was moderated by individuals' income level and their perceptions of formal and informal institutions.

Their results show that participation in the shadow economy in a transition country like Albania is indeed related to individuals' education level. The more highly educated the individuals, the less likely they are to participate in the shadow economy. However, the financial argument that the more highly educated have more opportunities and higher income and thus they have less reason to engage in the shadow economy, explains little. If anything, education seems to affect people's lower participation in the shadow economy through the encouragement of more civic social norms and higher tax morality. The authors conclude that, nevertheless, more research is needed to understand how education relates to the shadow economy because the channels they investigate do not fully explain the relationship.

Krynska et al. (2007) use quantitative and qualitative evidence and make two important observations regarding education and informal work in Poland. First, the informal labour market is a segmented market consisting of: (a) an upper tier where highly educated people engage in informal work not out of necessity, as most of them have a job in the formal sector, but rather to evade taxes for financial gains, and (b) a lower tier where people with lower levels of education engage in informal work to make a living because they are not able to find a job in the formal sector. Second, people with higher education engage more in a certain type of informal employment, namely quasi-formal employment,⁸ than those with lower educational levels (Krynska et al., 2007).

In two follow-up studies, Cichocki and Tyrowicz (2011) and Liwiński (2022) use the Polish Labour Force Survey data and find that those working informally are in general less educated than formal workers. Their findings confirm that the Polish informal labour market is a two-tier market, where workers in the lower tier have lower educational attainment, occupy unskilled jobs and earn less than formal workers, whereas workers in the upper tier are better educated, have more skilled jobs and earn similar wages to those working in the formal sector.

2.4 Trust and the shadow economy

Does individuals' involvement in the shadow economy depend on the trust-based exchange with the government and amongst themselves? Take, for instance, tax evasion; standard economic theory would predict that individuals will pay taxes if they are forced to through audit and penalty. Public choice theory, on the other hand, would argue that if people feel that paying taxes is ‘rewarded’ by governmental services, they would evade less (Friedman et al., 2000). As such, tax compliance can be regarded as a social contract between citizens and government with the quality of public services being the key factor (Rothstein, 2005, 2021; Uslaner, 2010). Yet, free-riding can endanger the exchange between citizens and government, because individuals would rather benefit from public goods without having to pay for them. Free-riding may be mitigated, however, if individuals trust that their government will fulfil its promises and that other individuals will do their share and pay the taxes. Indeed, various studies report a negative relationship between trust and the shadow economy, especially in developing and transition countries. For instance, D'Hernoncourt and Meon (2012) find that both generalized trust in others and trust in government are negatively associated with the size of the shadow economy; Amini et al. (2022) observe that trust in state institutions increases people's willingness to report corruption; and Polese et al.'s (2022) study of managers in Kyrgyzstan in 2017–2018 confirms that bureaucracy, corruption and a lack of trust in state support and institutions increase participation in the shadow economy. This alternative perspective relates to the concept of a psychological contract (Feld and Frey, 2002) between the taxpayer and the government. The basic idea behind such a contract is that a mutual exchange requires that tax authorities act respectfully towards taxpayers, which in turn may lead to more trust and compliance from the citizens. In fact, Easter (2002) reports that this is what happened in Poland in the early 1990s, when the government earned the trust of its citizens by being honest.

An extensive literature on tax morale⁹ has emphasized the importance of trust in formal institutions with regard to tax morale (e.g. Alm and Torgler, 2006) and especially so in developing countries (e.g. Torgler, 2005). In line with this literature, Gërxhani and Wintrobe (2021) try to establish a link between new institutionalism and the emergence of a mutual exchange between the taxpayer and the government. Under the assumption that informal institutions, as captured in trust, determine the extent of tax compliance, they first develop a theoretical model that combines the public choice approach with the new institutionalist perspective. Then they empirically test the model's prediction that an individual will pay taxes conditional on believing that their government will provide the services promised and that other individuals will pay taxes as well (i.e. trust-based exchange). The empirical test is based on the observational data collected in 2000 in Tirana, the capital of Albania (see Gërxhani, 2004b). Aside from a measure of tax evasion, the data contain information on individuals' perceptions of their government (e.g. related to corruption or the overall support it deserves) and the compliance level of their fellow citizens.

Their results support the model's prediction by showing that the more individuals trust the government and believe that others pay taxes, the more they are inclined to pay taxes themselves. These results remain strong after controlling for various socio-demographic characteristics. Education, however, somewhat mitigates the relationship between trust in the government and tax evasion because being highly educated is positively correlated with more trust in the government. This finding relates to the earlier study by Gërxhani and van de Werfhorst (2013) on the relationship between education and participation in the shadow economy and supports the suggestion of promoting more (civic-oriented) educational training if more compliance and a smaller shadow economy is desired.

2.5 Opportunities to evade and the shadow economy in a comparative setting

If both formal and informal institutions matter for individuals' participation in the shadow economy, an interesting question relates to the relative importance of the two. This question is difficult to answer with observational data because they typically occur in combination, making it challenging to isolate the causal effect of either type of institution in itself. To address such a challenge, Gërxhani and Schram (2006) conduct a comparative experimental study of tax evasion in Albania and the Netherlands. The motivation for comparing a transition country with a developed nation is twofold: (1) the magnitude of the shadow economy is estimated to be much larger in Albania than in the Netherlands (roughly 40–50% and 10–20%, respectively), and (2) their institutional setting is also very different, with the Netherlands having a long-standing tradition of well-functioning formal institutions. The authors argue that in line with the new institutionalist approach, it may seem straightforward to conclude that the smaller shadow economy in the Netherlands may be because of its well-functioning formal institutions. Yet, because informal institutions may be equally important, it may also be the case that the Dutch have a higher tax morale than the Albanians.

To disentangle the effects of formal from informal institutions, Gërxhani and Schram (2006) conduct a controlled experiment, in each country, that implements exactly the same rules (formal institutions) under which the participants' tax evasion behaviour is observed. The rationale behind this design is that if the Dutch and the Albanians show differences in their tax evasion behaviour, the authors can then confidently say that these differences are not driven by formal institutions. Aside from a cross-country comparison, the experiments also test whether the decision to evade taxes differs across socio-economic groups within each country. Importantly, the experimental design systematically varies the opportunity to evade taxes and the audit probabilities, while also including a measure of individuals' risk attitudes.

More specifically, the experiments in both countries took place in 1999–2000 with high school pupils, university students, high school teachers, university non-academic personnel, and university teachers, in Albania and in the Netherlands. Before deciding to evade taxes or not, each participant chooses between two types of income, knowing that one type gives an opportunity to evade while the other does not. After this choice, an income of that type is randomly determined and privately made known to the individual. Subsequently, participants report an income to be taxed, whereby those with an ‘evadable’ income type can report less than their actual income. Their reported income can be audited, which happens with a probability of either 1:2 or 1:6 (depending on the treatment). Those caught evading taxes are fined. In most sessions, tax proceeds are redistributed to the participants as a public good. For comparison, some subject pools participate in additional sessions where there is no such redistribution.

The main findings of the study are that Dutch participants evade taxes more than Albanian participants. Across the different socio-economic groups, pupils and students in each country evade taxes the most. This is an indication that within-country differences in tax evasion behaviour are as relevant as cross-country differences. As expected, when given a choice, participants who are more inclined to evade taxes prefer labour market conditions (i.e. an income type) that allow for more opportunities to do so. Interestingly, Albanians' tax evasion behaviour is not affected by the chances of being caught, while the Dutch evade taxes more when the probability of being audited is lower. In fact, not only do Albanians have an overall higher compliance rate, but they even over-report (i.e. pay more taxes than required) more frequently than the Dutch. However, this distinct behaviour of Albanian participants is observed much less frequently in the sessions without the redistribution of the tax proceeds.

The authors conclude that contrary to what the aggregate data show about the shadow economies in Albania and the Netherlands, the opposite is observed when holding formal institutions constant: Dutch participants evade taxes more than Albanian participants. They argue that the experimental difference cannot be attributed to formal institutional differences, because the experimental formal institutions are the same in both countries. The question, then, is how much can be attributed to differences in social norms that are internalized in participants' behaviour in the experiment? Because Gërxhani and Schram (2006) do not investigate the direct effect of norms, one can only speculate based on their findings that Albanians evade less, that their compliance does not depend on audit probabilities, and that they over-report more frequently than the Dutch. All of these differences are especially pronounced when tax proceeds are redistributed amongst the participants. Because informal institutions like norms take time to change, it is likely that the legacy of past communist norms of common property and sharing with others are still ingrained in Albanians' behaviour of contributing to an experimentally well-functioning and enforced collective good. Conditional on the latter, it is possible that this willingness to contribute provides an explanation for all the experimental differences that are observed between the two countries. The within-country differences where the younger generation (i.e. pupils and students) evades more than the rest, reinforces this proposition.

Yet, there are possible alternative interpretations of the findings obtained by Gërxhani and Schram (2006). First, the perceived probability of being audited could differ among the Albanians and the Dutch. It is possible that it is the perception and not the probability of being audited per se that have led to different compliance rates in the two countries (Bergolo et al., 2017; Mascagni, 2018). Second, reputation, social stigma and public shame may also play a role in explaining the differences. Coricelli et al. (2010) and Coricelli et al. (2014) show that tax compliance may be driven by the willingness to avoid negative emotions raised by potential detection and public denouncement. Also Alm et al. (2017) find that public shame seems to be a powerful driver of tax compliance. Finally, social pressure can be another explanatory factor. As documented by Battiston and Gamba (2016), direct peer pressure can lead to a rise in tax compliance.

3. Effects of participation in the shadow economy

3.1 The shadow economy and well-being in Albania

How does participating in the shadow economy relate to individuals' economic well-being, and what explains it? Compared to working in the formal sector, research shows that participation in the shadow economy offers the obvious advantage of not having to pay taxes, of often facing fewer (bureaucratic) barriers to entry and of having more flexibility and creativity. Instead of ‘doing their business’ in the official economy, individuals move into the shadow economy where they do not have to interact with state institutions and do not have to comply with burdensome regulations. This is known as the ‘exit’ perspective (e.g. Maloney, 2004; Perry et al., 2007; Williams, 2010a). On the other hand, those involved in the shadow economy have, on average, lower earnings than those working in the formal sector, typically face poorer working conditions and higher job insecurity, and are excluded from social benefits because of working without a labour contract. Such kind of employment is ‘sought’ by those who cannot find any other possibility of employment – be it because of their education, qualifications, experience or lack of opportunities. Because formal employment is unattainable for them (Williams, 2010a), they face the choice of being either inactive or unemployed, or of trying to survive by engaging in the informal sector. The latter situation as a survival tactic and a substitute for formal work is known as the ‘exclusion’ perspective (Banerjee, 1982; Perry et al., 2007; Williams, 2010a).

Various studies have suggested that the exclusion perspective is more relevant in transition countries (e.g. Williams, 2014). With an economic crisis and high unemployment rates in the post-socialist period, there were limited formal sector jobs available and, moreover, there was only meagre, if any, financial support by the welfare system. This suggests that involvement in the shadow economy in transition countries may not be a voluntary choice but more of a survival mechanism as a result of being ‘excluded’ from formalized employment alternatives and their benefits (e.g. Gërxhani, 2004a; Schneider and Enste, 2000).

Against this background, Ferrer-i-Carbonell and Gërxhani (2011) tries to empirically investigate how participating in the shadow economy of a transition country relates to individuals' economic well-being. The authors focus on a broader concept of economic welfare, which is captured by individuals' own evaluation of their financial situation. In light of the (dis)advantages of participating in the shadow economy, the authors argue that when studying the effects of such participation, one should consider a concept of welfare that measures more than just income. For instance, individuals involved in the shadow economy may feel less satisfied with their financial situation (even if they earn a decent income) because of financial uncertainty, social insecurity or having been given a bad reputation and low status. Because individuals are best to judge their own financial situation, the authors apply a subjective method and ask: ‘How satisfied are you with the financial situation of your household?’, which gives them an indication of a respondent's perceived economic situation and thus their subjective well-being. This method, eliciting scores on a scale from ‘0 = not at all satisfied’ to ‘10 = very satisfied’ is frequently used in the well-being literature to measure individuals' financial satisfaction (e.g. van Praag and Ferrer-i-Carbonell, 2004).

Ferrer-i-Carbonell and Gërxhani (2011) argue that if the disadvantages are perceived to outweigh the advantages, one could expect individuals involved in the shadow economy to feel less satisfied with their financial situation than individuals in the formal sector. This effect may be especially strong in transition countries where participation in the shadow economy is more likely a survival mechanism due to the lack of formalized alternatives. Moreover, the authors expect differences among individuals participating in the shadow economy according to their social norms, because individuals' satisfaction with their financial situation may depend not only on their economic state but also on their social context. Since paying taxes or not has important consequences for the own and public finances, the authors consider two dimensions of tax morale. The first relates to how individuals perceive the financial consequences of (not) paying taxes and the second dimension refers to beliefs about contributing to public goods by paying taxes. These dimensions are expected to moderate the relationship between individuals' participation in the shadow economy and their financial satisfaction.

To test these propositions, Ferrer-i-Carbonell and Gërxhani (2011) use the observational data collected in Tirana, Albania (see section 2.1), which include information on individuals' financial satisfaction (average: 3.37 out of 10), their informal sector participation (as defined in section 2.3), their household monthly income and taxes, their tax morale (based on their attitudes towards two statements related to paying taxes) and a set of controls on their socio-economic and demographic attributes such as age, education, gender, etc.

The results indicate that on top of the differential income effect, most individuals who work in the informal sector are less satisfied with their financial situation than those with formal employment. There is, however, a small group of individuals who appear to be more satisfied with being involved in the informal rather than the formal sector. The main characteristic of this group is a low tax morale; that is, they do not perceive severe financial consequences in the future of not paying taxes today. It is possible that these respondents are less risk-averse than others or have economic buffers that allow them to worry less about social insecurity or financial uncertainty than others do. The majority of respondents, however, do seem to share concerns about future financial insecurities and are thus less satisfied with working in the informal sector and not paying taxes. The authors argue that this was perhaps to be expected in a transition country like Albania, where a lack of formal employment possibilities and trivial social support may ‘force’ many people to engage in the shadow economy while not being happy about this ‘choice’.

However, some studies point out that neither the ‘exit’ nor the ‘exclusion’ perspectives can sufficiently explain informal work participation, because there is considerable heterogeneity regarding types of informal work, which also ‘mingle’ with formal work (Bargain and Kwenda, 2014; Fields, 2005; Gunther and Launov, 2012). This can lead to the emergence of different types of hybrid work and a ‘blurring’ of borderlines between formal and informal employment (Williams, 2010b). Hence, these studies propose that a combination of both perspectives may apply: informal work is for some people an attractive employment opportunity whereas for others it is the only possibility of employment. Indeed, empirical studies show that informal labour markets are divided into two parts: (a) an upper tier, which represents the competitive part of the informal sector and where wages are comparable to those in the formal sector, or (b) a lower tier, dominated by those for whom formal employment is unattainable and where a significant wage penalty is present (Bargain and Kwenda, 2014; Gunther and Launov, 2012). Such a two-tier structure has been confirmed by Liwiński (2022) in the case of Poland: informal workers in the first quartile of the wage distribution and those above the first quartile are in two partially different segments of the informal labour market. The former have substantially lower monthly earnings than formal workers, while the latter have the same monthly earnings as formal workers.

3.2 The shadow economy and well-being in 14 transition countries

The above discussion raises an important question. The study by Ferrer-i-Carbonell and Gërxhani (2011) suggests that their findings on the relationship between participation in the shadow economy and individuals' welfare may indicate that the exclusion perspective is more relevant than the exit perspective in transition countries. On the other hand, other studies on transition countries argue that both perspectives may be relevant. The question thus is whether the negative well-being effect observed in one country holds for a broader set of transition countries. To address this question, Ferrer-i-Carbonell and Gërxhani (2016) conduct a similar follow-up study based on data from 14 transition countries. Given the moderating effect of tax morale in their study on Albania, the authors examine the role of the social and institutional context in shaping the relationship between tax evasion and well-being. In doing so, they focus on a richer set of factors, including tax-related formal and informal institutions as well as formal and informal types of social capital.

Their rationale for considering tax-related formal institutions is based on the finding that in transition countries tax evasion is positively related to ineffective formal institutions because the latter lead to a decline in civic virtue and in loyalty towards public institutions (e.g. Schneider and Enste, 2000; Torgler, 2012). Thus, for those individuals who think that formal institutions do not do a good job, the authors argue that the negative relationship between evading taxes and well-being will be weaker. As for the moderating role of informal institutions, they expect that because individuals with positive tax-related social norms are generally happy (Lubian and Zarri, 2011) and are less likely to evade taxes (e.g. Cummings et al., 2009; Torgler, 2003), they will experience a stronger negative relationship between tax evasion and well-being.

When looking at the role of social capital, Ferrer-i-Carbonell and Gërxhani (2016) argue that the more formalized types of participating in associations (e.g. membership, volunteering or donating; Putnam, 1993) contribute to a sense of civic responsibility towards the society. In turn, this civic responsibility is predicted to decrease non-compliance behaviour (Letki, 2006; Orviska and Hudson, 2002). They do, however, make a distinction between Putnamesque and Olsonian associations. The former promote common goals without imposing negative externalities on the rest of society, while such externalities do occur more often with the latter (Fidrmuc and Gërxhani, 2008; Knack and Keefer, 1997). Thus, the authors expect that those participating in Putnamesque associations care more about their civic duty and will therefore experience a stronger negative relationship between evading and well-being. In contrast, those participating in Olsonian ‘rent-seeking’ associations will be less concerned with their civic responsibility and thus experience a weaker negative relationship – that is, they will feel less dissatisfied with evading taxes.

Finally, they also consider the role of social capital in its more informal form through networks of family and friends. Rose (2000) finds a positive relationship between such networks and underground activities in Russia. He explains this by the malfunctioning of formal institutions, which ‘forces’ people to fall back on their networks and bypass the state. The authors thus argue that having networks of family and friends may moderate the negative relationship between evading taxes and well-being. This moderating effect, however, may depend on the willingness of network members to pay taxes and the degree to which they put social pressure on others to comply with their tax-related social norms. Based on this reasoning, the authors predict that having networks of friends and family with high (low) tax morale would strengthen (weaken) the negative relationship between evading and well-being.

Ferrer-i-Carbonell and Gërxhani (2016) test these hypotheses by using observational data collected as part of the ‘Public Goods through Private Eyes’ survey, carried out in 2013 and 2014 by the University of Warsaw (Letki, 2015). The response rate in most countries is around 45–50%. The objective of the survey is to collect cross-country data on the determinants of public attitudes and behaviour towards public goods in the context of post-communist Central and Eastern Europe. The fieldwork is carried out in 14 countries.¹⁰ The final sample consists of more than 22,000 observations and each country has at least 1,400 observations.

The dependent variable measuring subjective well-being is obtained by asking people about their individual life satisfaction, i.e. ‘All things considered, how satisfied are you with your life as a whole nowadays?’. Similarly to the financial satisfaction question used in the survey in Albania (cf. section 3.1), the answers are given on a scale from ‘0 = completely dissatisfied’ to ‘10 = completely satisfied’. The independent variable measuring tax evasion is obtained by asking respondents whether they have ever worked for cash in hand and not paid tax on the income. The institutional moderating variables are measured through people's attitudes on a number of statements. Specifically, individuals' perception of tax-related formal institutions is captured by ‘individuals' positive perception of formal institutions’ and ‘individuals reporting underprovision of public services by the state’, while tax-related informal institutions are captured by ‘individuals' high tax morale’. The formal type of social capital is measured through ‘individuals' participation in civic associations via financial donations’ (Putnamesque type) and ‘individuals' participation in rent-seeking associations via financial donations’ (Olsonian type). Finally, the informal type of social capital is captured by the ‘frequency individuals meet with people (family, friends, colleagues, neighbours)’.

Similarly to the Albanian case (section 3.1.), their results show a strong negative relationship between tax evasion and life satisfaction. As predicted, this relationship varies depending on the social and institutional context individuals are embedded in. The conclusions from their regression results are summarized in Table 2. When considering the role of formal and informal institutions, only negative experiences with tax-related formal institutions (i.e. underprovision of public services) have a significant effect; that is, such negative experiences weaken the negative relationship between tax evasion and life satisfaction. This is in line with findings from other key studies on happiness in transition countries, which conclude that people in these countries suffer from a ‘happiness gap’ because of the deterioration of public goods (Guriev and Zhuravskaya, 2009), the malfunctioning of formal institutions (Nikolova, 2016), higher perceived corruption and weaker government performance (Djankov et al., 2016), among other factors. As for social capital, the authors observe that only the formalized type of participating in both types of formal associations has a significant effect. That is, individuals supporting civic associations feel a stronger negative relationship between evading taxes and their life satisfaction, while for those supporting ‘rent-seeking’ associations the negative relationship becomes weaker.

Table 2. Moderating effects of social and institutional context on the relationship between tax evasion and life satisfaction

In an explorative analysis, Ferrer-i-Carbonell and Gërxhani (2016) run the same (regression) analysis separately for those who had had working experience during the communist period (45 years or older in 2013–2014) and those who had not (younger than 45 years old in 2013–2014). The idea underlying this exercise is to explore the effects of having internalized social norms and experiences from the communist past, which they argue is only to be expected in the 45+ group. The results show that both generations experience a weaker negative relationship between tax evasion and satisfaction when they believe the government is underproviding public services. However, the two generations differ in how their social capital affects them. While the younger generation is more engaged with associations and experiences a stronger (weaker) negative relationship when participating in civic (rent-seeking) associations, the older generation is more dissatisfied with evading taxes the more frequently they meet with their family and friends.

The authors conclude that, generally speaking, individuals participating in the shadow economy in transition countries are not happy with evading taxes. Thus, the negative well-being effect observed in one country seems to hold for a broader set of transition countries, which again hints at the exclusion perspective being more relevant in these countries. This depends, however, on the social and institutional context one lives in. ‘While for some individuals tax evasion is not negatively associated with life satisfaction, the way they experience various social and institutional factors turns this relationship into negative. This implies that if these individuals had a choice, they would rather enter a formalized labor market’ (Ferrer-i-Carbonell and Gërxhani, 2016: 157–158). The moderating effect of the current context and past legacy is especially noticeable when examining the relationship between tax evasion and individual life satisfaction across generations. While the tax-related behaviour of the older generation seems to be more affected by the past through social networks one is embedded in, that of the younger generation is more linked to their association engagement during the transition period.

4. Conclusion

The research summarized in this paper shows that the shadow economy in transition countries remains a challenge not only due to its large magnitude and the implications for state budget deficits, but especially due to its ‘survival’ or ‘mingled’ nature and the implications thereof for individuals' well-being and for their exchange with formal institutions, civil society and social norms.

The complementarity of the distinct research designs applied in the studies overviewed enables a good balance between the internal and external validity of their empirical findings. For instance, in the study where formal institutions and their enforcement were held experimentally constant across countries, Albanians evade taxes less than the Dutch. Yet, field evidence points to the contrary: the Albanian shadow economy is larger than the Dutch shadow economy. This means that while in the field one cannot distinguish whether this outcome is due to formal or informal institutions, the controlled experimental setting allowed the authors to observe that the Albanians evade taxes less often when formal institutions function properly and are enforced. By implication, one possible interpretation of this result is that Albanians may not have a low willingness to pay taxes (i.e. informal institutions). The control ensured in an experimental setting enables a better understanding and interpretation of the field evidence, which is why in our view more experimental research is needed to complement observational evidence on shadow economies.

The fact that the field evidence tells a different story than what was observed in the lab-in-the field experiment is thus less likely related to informal institutions like low tax morale and more likely related to the malfunctioning of formal institutions, their poor enforcement or their interaction with informal institutions. The observational data in Albania confirm this claim because when individuals' perceptions of formal institutions clash with their informal institutions, individuals are more involved in the shadow economy. Instead, when people have a trustworthy relationship with the government and other individuals, they are more inclined to pay taxes. The importance of their social and institutional context also appears in how individuals' involvement in the shadow economy relates to their subjective well-being. When they do not believe the government is sufficiently providing public services or when they do not foresee any financial insecurity in the future from not paying taxes today, they are happy to evade taxes. The majority, however, are unhappy to be ‘stuck’ in the shadow economy and having to evade, and this unhappiness becomes stronger the more they are concerned about being excluded from social benefits or the more they are civically engaged and would rather contribute to the collective good.

This research was theoretically inspired by the new institutional perspective that a proper functioning of the economy needs the formal and informal institutions to be complementary and well enforced. Its main contributions are in empirically showing that: (1) focusing on formal institutions alone, that is strengthening the rule of law, is a necessary but insufficient response to the shadow economy; (2) taking account of informal institutions, such as individuals' tax morale or their trust in government and in other citizens' civic morality, is as necessary; and (3) most importantly, formal and informal institutions go hand in hand, and as such, their interaction ought to be an essential part of the new institutional perspective. In fact, we propose that the research looking at different facets of ‘culture’ and ‘rule of law’ can be understood within the new institutional framework and its relationship with economic development, engagement in informal activities and overall lower economic welfare.

Although the focus of this paper has been on transition countries in general, and Albania in particular as an ideal context of institutional analysis, the above contributions are globally relevant. As various studies show, the relationship between culture and institutions is also very important for the economic progress of developed countries. Yet, the jury is not yet out on whether institutions affect culture or vice versa. For example, Letki (2006: 321) proposes a ‘top-down mechanism, where institutional quality is enhanced and political confidence is stimulated, and in turn they generate respect for norms and rules and positive attitudes to compliance’. Tabellini's (2010) findings also indicate a positive effect of institutions on economic performance through culture. Gorodnichenko and Roland (2017), on the other hand, conclude that because culture changes more slowly, it can have an important influence on formal institutions, which in turn contribute to economic development.

This debate is, however, crucial for transition economies, which have undergone major (formal) institutional reforms. These institutions were often transplanted from established and well-functioning democracies to radically different cultural contexts. In light of the importance of the two-way interaction between culture and institutions in economic development in general, and of the importance of the two-way interaction between formal and informal institutions in the shadow economy in particular, we are now better able to understand what went wrong in transition economies and what can be done in the future.

First, the implementation of formal institutional reforms should go hand in hand with ways to align them with the cultural context. A recent study by Kaasa and Andriani (2022) shows that people have more trust in formal institutions the more they live in civically engaging communities. This is in line with some of the findings reported in this paper, where there seems to be a generational shift in the importance of the social and institutional context when judging whether being involved in the shadow economy increases one's satisfaction. While the older generation is still affected by the legacy of social norms during communism, the younger generation cares more about participating in associations and experiences a stronger negative relationship between tax evasion and happiness when participating in civic associations. Thus, alongside the strengthening of formal institutions, governments and civil societies in transition countries should also try to encourage people's civic responsibility and create opportunities for participation and engagement. The latter should also promote a more trust-based exchange between the government and its citizens, which may be a better response to engagement in informal activities than more coercive enforcement through audit and punishment.

Second, to encourage people's civic responsibility and engagement in civic associations, which seems to be more prominent among young people, a special role can be dedicated to the promotion of more civic-oriented educational training. In line with Letki's (2006) finding that education positively affects one's civic morality, the research presented here shows that higher education is positively related to more trust in government, but also to less involvement in the shadow economy.