Corruption problem or perception problem?

Corruption problem or perception problem?

Date: April 23, 2018
Corruption problem or perception problem?

By Jan Bednorz,
CRRC-Armenia Intern
To follow up on my previous post that covered the Armenian institutional framework in general, herein I elaborate specifically on the problem of corruption in Armenia and attempt to explain disparities between different measures of corruption. I once again use the results of the Life in Transition Survey (LiTS) and complement it with the Caucasus Barometer data and some empirical studies on this topic.

Even though adverse effects of corruption are widely recognized and unquestionable, the nature and high complexity of the phenomenon make the absolute levels of corruption difficult to measure. Discrepancies between the two popular approaches: perception and experience of corruption are especially appealing. It may seem as a cause-effect relationship, but in fact, it is not the case in many countries, Armenia being one of the most striking examples.

Although this topic is not yet well-researched, some studies indeed point out that corruption experience is a weak predictor of corruption perception (Bohn 2013; Donchev and Ujhelyi 2014).  Instead, what appears to shape people’s corruption perception is their assessment of country’s economic development and key democratic institutions. So, if they perceive the country to be prosperous and the government inclusive and efficient, they also assume that public administration is fair and transparent.

I also believe that part of the explanation might be the nature of the questions asked in surveys that do not account for complexity of corruption. A question “Have you paid a bribe to a public official?” is pretty straightforward and measures specifically, so-called, petty corruption (bribery at the low end of politics when the public official meets the citizen). However, “Are bribes necessary across public services?” makes one think about the problem more broadly and consider a bigger picture of corrupted system and hence, might unintentionally involve other types of corruption: grand (at the top end of politics where the rules are formulated) and political (abuse of power, election rigging, etc.).

These assumptions hold true for the LiTS data – the correlation between perception and experience of corruption across countries is weak (r=0.41). Therefore, to check what might shape corruption perceptions, I tested the difference between levels of experienced and perceived corruption against some other indicators from LiTS database. I came to a conclusion that the disparity is best explained by the level of trust in the government (r=0.72; see Figure 1). Hence, the higher trust in the government, the more likely people are to underestimate the levels of corruption. Accordingly, if people do not trust the government, they tend to overestimate its extent.

What one might find also interesting about this figure, is the number of observations above the 0 line – the corruption experience is on average by 4.3 percentage points higher than perception, which means that in most countries people are excessively optimistic about the levels of corruption (perception). However, two striking examples of excessive pessimists can be noticed as well, one being Moldova and the other – Armenia (see red circle in Figure 1).

According to LiTS, 35% of Armenians say that unofficial payments or gifts are usually or always necessary in public services but only around 10% report having actually made any (the disparity is -25% points, see Figure 1). Similar discrepancies are observed in other datasets – according to Caucasus Barometer 2017, only 3% admits to have paid a bribe but in the same time, corruption is seen as one of the most important issues facing the country.

As argued above, this situation can be explained by very high levels of distrust and dissatisfaction with public institutions in Armenia (see my previous post) that result in inflated perceptions of corruption. Moreover, going back to my second assumption, even though the experience of (petty) corruption might be relatively low, Transparency International ranks Armenia very low in the Corruption Perception Index (107th place out of 180 countries assessed). They precisely call attention to strong patronage networks and the overlap between political and business elites (grand corruption) as well as a history of flawed electoral processes, including fraud and vote-buying (political corruption) (Transparency International 2013). If respondents indeed take these issues into account when asked about perceived corruption but not experienced corruption, the gap is understandable.

Worryingly, corruption is considered to behave as a self-fulfilling prophecy – if people believe corruption is widespread and thus, necessary in dealing with public administration, they are more likely to engage in bribery themselves (see, for example: Inter-American Development Bank 2015). This self-perpetuating vicious cycle goes beyond that though; governance deficiencies (including corruption) fuel and at the same time are fuelled by citizens’ dissatisfaction, political apathy and disbelief in any system change (also in fight against corruption). All these symptoms are clearly visible in Armenia. Since, without a strong bottom-up pressure, political actors have very little interest in improving state institutions or eradicating corruption, breaking out of this vicious cycle would require a vibrant civil society, independent and active media and, ideally, a fresh political class in a shape of effective and strong opposition.

  1. Bohn, S. R. (2013). Corruption in Latin America: understanding the perception–exposure gap. Journal of Politics in Latin America, 4(3), 67-95.
  2. CRRC (2017). Caucasus Barometer. Available at:
  3. Donchev, D., & Ujhelyi, G. (2014). What do corruption indices measure?. Economics & Politics, 26(2), 309-331.
  4. European Bank for Reconstruction and Development (2017). Life in Transition Survey. Available at:
  5. Inter-American Development Bank (2015). Corruption as a self-fulfilling prophecy: evidence from a survey experiment in Costa Rica. IDB Working Paper Series No. IDB-WP-546.
  6. Transparency International (2013). Overview of corruption and anti-corruption in Armenia. Available at:

The content of this blog is the sole responsibility of the author and does not necessarily reflect the views of CRRC-Armenia.