A recent column on (Slemrod et al. 2019) started with “tax evasion is a pervasive problem in developing countries and a non-trivial one in developed countries.” 

Tax evasion takes many forms. While its magnitude may vary significantly across places,1 almost all countries share a common interest in learning how to fight evasion. The urgency of tackling tax evasion grows in periods of economic crisis, when governments consider austerity measures or when tax fraud scandals (e.g. Panama Papers or Germany ex/cum trades) come to public attention and, inevitably, enter the political debate. Although the legal and judicial system may provide the right incentives to prevent evasion, tax agencies and their auditing teams play a crucial role in identifying and punishing offenders.

Tax fraud in the real estate sector

While taxes on property transactions are common in most OECD countries, they remain understudied (Best and Kleven 2018). In 2006, OECD’s Centre for Tax Policy and Administration began a systematic and comprehensive examination of tax fraud and money laundering which also reported on such vulnerabilities in the real-estate sector. 

The report surveys 18 countries in order to assess the extent of illegal practices in different countries and sectors. The published document (OECD 2007) reports that “in most of the countries surveyed, the real-estate sector has been identified as an important sector being used to facilitate tax fraud and money laundering. There are no reported official figures or statistics about the dimension of the problem; one country, Austria, has reported an estimation of the amount of fraud involving the real-estate sector of approximately €70 million.” Price manipulation (escalating prices make it easier to manipulate prices of properties and transactions), undeclared income/transactions and the use of false identities are found to be the most common methods used when committing tax fraud.

How can tax fraud be limited by policy?

Auditing is one potential option. However, while it is not only an extremely costly measure it also requires information on taxpayers and their activities that is often difficult if not impossible to collect. Tax authorities, in most cases, either randomly select transactions to be audited or they use a red flag system. Such a system typically chooses transactions to be audited based on risk profiling, data matching, statistical analysis or data mining. Choosing the best red flags is crucial in order to manage the tax authority’s resources efficiently. A good red flag must be easy to observe and the red flag system must exhibit a high probability of identifying fraud.

Appraisals as an instrument to capture transfer-tax evasion

In most Western countries, an official appraiser must estimate the value of a property if a potential buyer asks for a mortgage. There is evidence that the appraisal often overvalues the property: in the US the appraised price is (weakly) larger than the transaction price in more than 95% of cases and an increase in inflated transactions was observed between 2000 and 2006 (Cho and Megbolugbe 1996, Nakamura et al. 2010, Ben-David 2011). The figures on inflated transactions are found to be even higher for Spain (Montalvo and Raya 2018). Interestingly, the UK is one of the countries with the highest accuracy of appraisals (Cloyne et al. 2019) while exhibiting the lowest evasion of the stamp duty land tax (Best and Kleven 2018).

READ ALSO  SE: H.K.-Singapore Travel Bubble Is Major Milestone: Tourism Board's Cheng

In a new paper (Montalvo, Piolatto and Raya 2020), we study the extent to which appraisals can be used as an instrument to capture the likelihood of transfer-tax evasions. We develop a theoretical model and combine it with an empirical analysis on Spanish data from 2005 to 2011.

Our model is based on the well-known deterrence model (Allingham and Sandmo 1972) where the threat of detection and punishment transforms the choice of potential evaders into a lottery where they weight the expected cost and benefit of evading, depending on the probability of being audited. In our model, the mortgagor would like to minimise the expected transfer-tax burden by optimal under-reporting of the sales value. However, under-reporting requires having enough liquidity for the down payment and the (illegal) side payment. On the other side, the mortgage can only pay part of the expenditure, computed as a share of the property appraisal. The mortgagor and the mortgagee can however collude in order to inflate the value of the appraisal and hence the final mortgage, which relaxes this constraint.2 

Depending on a buyers’ access to liquidity, on one extreme there are constrained agents who are unable to cheat but who would like to inflate the appraisal. On the other extreme, there are unconstrained agents, with no need to inflate the appraisal, who are willing to under-report the sale value. 

While the tax authority cannot observe the real transaction price, both the declared price and the appraisal are publicly disclosed and can easily be collected by the tax authority. We then build an index of observed over-appraisal, which is the ratio between the declared appraisal and the declared sale value, and we study how over-appraisal co-varies with tax evasion. 

The model unequivocally points to a negative correlation between over-appraisal and evasion which operates through liquidity (or other hidden resources): when liquidity increases, the opportunity to evade increases, but the need for inflating the appraisal decreases.

READ ALSO  Words versus deeds: How central banks manage expectations

We test our prediction using a unique Spanish dataset3 which includes the sales value declared to the tax authority, the amount effectively paid, and the appraisal value for about 1,400 transactions. Close to half of the transactions in our sample include some undeclared amount, on average about 7.64% or 13,847€ of the total. Conditional on fraudulent behaviour, this percentage rises to 15.1% (€27,409). Figure 1 presents the share of undeclared money, conditional on evasion. For example, the percentage of undeclared money was above 20% in 24% of the fraudulent transactions.

Figure 1 Share of undeclared money, conditional on evading

We estimate the probability of a fraudulent transaction, as well as the determinants of the amount of undeclared money. We show that tax evasion and over-appraisal are strongly negatively related in all our specifications. In our preferred specification, an increase in over-appraisal by one point (that is, the appraisal value doubles the transaction price) coincides with a decrease in the probability of fraud by 66.5% and a decrease in the amount that remains undeclared of 26,660 euros.

The use of over-appraisal to increase the amount to be borrowed remains the ultimate recourse for a buyer, used only in case of no alternatives. Hence, over-appraisal is a signal of liquidity constraints, and it is unlikely to occur for agents who have liquid savings that can be used for side-payments.  Since declared over-appraisal is much easier to assess and observe than access to liquid savings or fraud, it should be used as an indicator for the likelihood of fraud. In particular, tax authorities should focus their audit efforts on transactions where the appraisal is relatively low.

Combining our data with indicators of local corruption, quality of local government, shadow economy and social capital, we can further conclude that part of the variation in the likelihood of committing fraud and its level is explained by the local environment. Furthermore, as we have access to individual characteristics for one-third of the sample, we can document that educational attainment is a good predictor of low levels of evasion.

Our study thus confirms that both trust in society and education play an important role for the level of fraud. Long-run policies may therefore use these channels to increase compliance. Furthermore, we show that tax authorities can increase their auditing performance by targeting transactions characterised by appraisal values that are low compared to the declared price.4

READ ALSO  Self-preferencing by gatekeeper platforms: Implications for digital regulation


Ben-David, I (2011), “Financial constraints and inflated home prices during the real-estate boom”, American Economic Journal: Applied Economics 3: 55–87.

Best, M C and H J Kleven (2018), “Housing market responses to transaction taxes: evidence from notches and stimulus in the U.K.”, The Review of Economic Studies 85(1): 157–193.

Cho, M and I F Megbolugbe (1996), “An empirical analysis of property appraisal and mortgage redlining”, The Journal of real-estate Finance and Economics 13: 45–55.

Cloyne, J, K Huber, E Ilzetzki, H Kleven (2019), “The effect of house prices on household borrowing: a new approach”, American Economic Review 109(6): 2104–2136.

Dwenger, N, H Kleven, I Rasul and J Rincke (2016), “Extrinsic and intrinsic motivations for tax compliance: Evidence from a field experiment in Germany”, American Economic Journal: Economic Policy 8(3): 203–32.

Medina, L and F Scheneider (2017), “Shadow economies: what did we learn the last 20 years”, IMF Working Paper WP/18/17:1–59.

Montalvo, J G and J M Raya (2018), “Constraints on LTV as a macro-prudential tool: a precautionary tale”, Oxford Economic Papers 70:821-845.

Montalvo, J G, A Piolatto and J M Raya (2020), “Transaction-tax evasion in the housing market”, forthcoming in Regional Science and Urban Economics.

Nakamura, L (2010), “How much is that home really worth? Appraisal bias and house-price uncertainty”, Business Review, 11–22.

OECD Centre for Tax Policy and Administration (2007), “Report on Tax Fraud and Money Laundering Vulnerabilities Involving the real-estate Sector”.

Traxler, C (2010), “Social norms and conditional cooperative taxpayers”, European Journal of Political Economy 26: 89–103.

Slemrod, J, O Rehman and  M Waseem (2019), “Pecuniary and non-pecuniary motivations for tax compliance: Evidence from Pakistan”,, 15 May.

Slemrod, J and C Weber (2012), “Evidence of the invisible: Toward a credibility revolution in the empirical analysis of tax evasion and the informal economy”, International Tax and Public Finance 19: 25–53.


1 Estimating tax evasion represents a challenging task. Interested readers may refer to Medina and Scheneider (2017) for a recent attempt to quantify it across countries.  Slemrod and Weber (2012) presents an analysis of the limits to the empirical study of tax evasion.

2 Recent advances in behavioural economics proved that pro-social behaviours, the ‘warm-glow effect’ and feelings of stigma are extremely helpful to reconcile theoretical predictions and data (Traxler 201, Dwenger et al. 2016). The model reflects such advances, including a stigma component (conditional on being caught) and a shame component (unconditional on it).

3 We constructed the dataset by merging information from several sources.

4 This information can still be combined with other elements that the tax authorities may have access to. However, the advantage of using appraisal versus other red flag signals is that we proved its reliability and that it is easy to collect.