Tax evasion is a pervasive problem in developing countries and a non-trivial one in developed countries. The standard positive model of evasion, the deterrence model (Allingham and Sandmo 1972), suggests that evasion is constrained by the threat of detection and punishment, and thus the standard policy approach to reducing tax evasion is either increasing the odds of catching tax evaders or increasing the punishments levied on evaders when they are caught.
However, tax evasion may also be constrained by social and psychological factors (Andreoni et al. 1998). Some individuals may feel guilt or shame from evading or pride from fulfilling their civic duty, while others may be influenced by the possibility of approval or sanctions from peers (Luttmer and Singhal 2014). While there is substantial empirical support for the deterrence model (Slemrod, forthcoming), there is still limited evidence on the importance of these social and psychological factors and whether governments can prime them for resource mobilisation.
To learn more about how policy impacts tax evasion, we analyse two Pakistani tax programmes whose effectiveness depends on deterrence as well as social and psychological motivations for compliance (Slemrod et al. 2019).
In the first of these programmes, the government began revealing the amount of income tax paid by every taxpayer in the country. The public disclosure programme was instigated by a series of press reports documenting that the majority of the country’s lawmakers had not been fulfilling their tax obligations. The programme began in the 2012 tax year and has continued since then. Each year, two tax directories are published, one for the members of parliament (MPs) and one for all taxpayers. The directories are available online in a searchable PDF format that can be downloaded freely by anyone.
Although the disclosure is universal, the intensity of the disclosure is not equal for everyone. The directory for general taxpayers reveals the name, a privately known numerical tax identifier, and the tax paid by each taxpayer. The only publicly-disclosed information that can link an observation in the directory to a particular taxpayer is the name.
Because of Pakistani naming conventions, it is quite common for people to have the same full name. For example, the most frequent name in the tax directory, Muhammad Aslam, appears 15,598 times in four years, with a typical year’s directory containing more than 60 pages listing the name Muhammad Aslam alone. On the other hand, about one-third of taxpayers have unique names.
This variation in name commonness implies that the intensity of the disclosure varies considerably across individuals, depending upon how common their name is. Taxpayers with very common names enjoy virtual anonymity in the disclosed records, while uniquely-named taxpayers are exposed perfectly. We exploit this variation in disclosure intensity in our empirical strategy, comparing the change in tax payments before and after the disclosure regime across taxpayers with frequent and unique names.
The Taxpayers Privileges and Honour Card programme
The second programme publicly recognises and rewards top taxpayers of the country. The Taxpayers Privileges and Honour Card programme began concurrently with the public disclosure programme. It acknowledges the top 100 taxpayers in each of four categories – self-employed individuals, wage-earners, partnerships, and corporations – and grants them certain privileges. The Honour Card holders are invited to a special ceremony hosted by the prime minister each year to “recognise their services to the nation”, as well as to the state dinners held on Pakistan Republic Day. In addition, they are eligible for benefits such as fast-track immigration and gratis passports.
As the Taxpayers Privileges and Honour Card programme applies only to the top 100 taxpayers of each category, we leverage this cut-off in programme eligibility to estimate its impacts. If social recognition and related benefits offered by the programme are valued, taxpayers close to the eligibility cut-off will increase their tax payments in order to remain in, or enter into, the top-100 group. We test this by comparing the yearly growth in tax liability reported by agents close to the cut-off with that of other top taxpayers.
Effects on tax compliance
These programmes can influence tax compliance through a number of channels. The disclosed information can expose an individual as a tax cheat if their tax payment does not conform to the level of income, consumption, or wealth observed by neighbours, friends, and other peer networks. The information can encourage whistleblowers to come forward, increasing the threat of being caught and restraining evasion through the deterrence effect. The shame and guilt resulting from the disclosure can also induce greater tax compliance. On the other hand, the programmes may stimulate feelings of pride and positive self-image if an individual is revealed to be a compliant or top taxpayer.
We combine the disclosed data for the years 2012–2015 with 2006–2012 administrative tax return data to create a long panel of tax records for 2006–2015. We document three key findings.
Figure 1 Intensive margin response to the public disclosure programme
Notes: The figure plots the difference-in-differences coefficients and 95% confidence intervals from the event study on a balanced panel of self-employed taxpayers, comparing taxpayers with unique names (name appears at the most ten times in the four years’ disclosure data) versus others.
- First, the disclosure of tax information induced a substantial response from taxpayers with less common names.
Figure 1 shows that the tax liability reported by taxpayers with less common names, who are thus more intensely treated by the disclosure, on average increased by around 9 log points compared to those with common names. Consistent with our expectations, the estimated effect changes directly with the programme intensity. It is strongest for the most uniquely named individuals, declines as name commonness increases, and becomes insignificant when the name-frequency approaches 300 (i.e. the name of the taxpayer appears at least 300 times in the four years of disclosed data). Along the extensive margin, the programme caused an increase by one to two log points in tax filing by individuals with less common names relative to others.
- Second, the disclosure had a far stronger impact on MPs.
The tax liability reported by them surged by more than 40 log points, and Figure 2 shows that their tax filing rate jumped up by around 60 percentage points, from around 30% to more than 90%.
Figure 2 Extensive margin response to the public disclosure programme – MPs
Notes: The figure plots the fraction of MPs who file their tax return in the year indicated in the horizontal axis. Vertical lines demarcate the time from which the public disclosure begins to have an effect on the tax filing of MPs.
The stronger response from MPs is not surprising. They are likely to be more sensitive to the revealed information because, in addition to inducing shame and guilt, it can reduce their re-election probability. The disclosure was also more salient for them. They were explicitly identified in the disclosed records through their constituency numbers, and the media was more likely to pick on their noncompliance; the name commonness of the MPs had no effect on the response.
- Third, the Taxpayers Privileges and Honour Card programme also had a large impact.
In a sample containing top 1,000 taxpayers of each category, the tax liability reported by ranked taxpayers between 70 and 130 grew by nearly 17 log points faster than others because of the programme.
Shifting social norms towards compliance
The increase in tax compliance due to public disclosure could be due either to a perception that the chance of evasion being detected had increased or to a shift in social norms towards compliance. We find that among individuals with low tax payments, those who were in neighbourhoods with more compliant individuals increased their tax payments over time after the programme was initiated, compared to those who were in neighbourhoods with fewer compliant individuals. This catch-up pattern of response is consistent with a model in which the presence of compliant peers creates pressure on noncompliant ones to conform (Bursztyn and Jensen 2015).
Second, we investigate if the electorate rewarded or punished MPs in the next general election, held in July 2018, on the basis of their tax payments, which became public information from 2012. We find a strong, positive association between tax payment and electoral success. Taken together, the two pieces of evidence suggest that the public disclosure and Taxpayers Privileges and Honour Card programmes may have initiated some shift of the social equilibrium toward compliance.
Our results have important policy implications. The programmes we study cost little in terms of economic resources. Thus, to the extent that they are effective in influencing both private and social behaviour, they potentially offer a cost-effective complement to the standard measures governments undertake to deter tax evasion such as audits and information reporting requirements. Of course, any such policy needs to balance the pro-social impacts of public disclosure against concerns such as loss of privacy.
Allingham, M, and A Sandmo (1972), “Income tax evasion: A theoretical analysis”, Journal of Public Economics 1(3-4): 323–338.
Andreoni, J, B Erard and J Feinstein (1998), “Tax compliance”, Journal of Economic Literature 36(2): 818–860.
Bursztyn, L, and R Jensen (2015), “How does peer pressure affect educational investments?”, Quarterly Journal of Economics 130(3): 1329–1367.
Luttmer, E F P, and M Singhal (2014), “Tax morale”, Journal of Economic Perspectives 28(4): 149–168.
Slemrod, J (forthcoming), “Tax compliance and enforcement”, Journal of Economic Literature.
Slemrod, J, O Ur Rehman and M Waseem (2019), “Pecuniary and non-pecuniary motivations for tax compliance: Evidence from Pakistan”, NBER Working Paper 25623.