Corruption in Mongolia according to Transparency International

Corruption is one of the most prominent features cited in any overview of Mongolian politics, political risk, human development or investment potential. Yet, any hard evidence on the prevalence, extent and mechanisms of corruption are very hard to come by. One of the most internationally visible single piece of evidence in this regard is Transparency International‘s Corruption Perception Index. The 2012 iteration was released on Dec 5, 2012 and below I take a look at Mongolia’s ranking in and of itself and relative to other countries.

Methodological Note

First a few words on the Corruption Perception Index. First of all and very importantly, this is an index of the perception of corruption, not of corruption itself. Presumably this perception correlates highly with actual corruption, but perhaps perception is liable to change more quickly than actual practices, or may lag practices in some circumstances.

Secondly, as the CPI measures perception, it is survey-based and thus subject to the challenges that all surveys face in terms of the recruitment and representativeness of respondents, the translation of words and concepts across languages, etc. These challenges are more severe in a place like Mongolia where we can’t compare any survey data with some general social survey that would give us a better handle on population composition and demographic variables beyond what is offered in the census. This is a challenge that was laid bare in polling in advance of the 2012 parliamentary election, for example.

Importantly, the CPI is a composite index, i.e. it relies on scores/results of a number of other indicators. Beyond that it is actually somewhat difficult to figure out what exactly is going on in the calculation of the Index. The part that I haven’t quite figured out is how much of a judgement by individuals is involved as opposed to a purely mathematical process of aggregating and standardizing other indicators.

One the one hand, TI mentions the involvement of experts in various methodology notes, on the other hand, the Index is clearly described as an “aggregate index, which draws on relevant questions from a number of different data sources that capture business and expert views” (CPI Updated Methodology 2012).

As far as I can figure the methodology notes out, the basis steps in compiling the index are:

1. selecting data sources according to four criteria: quantifies perception of corruption in public service, valid methodology, credible source, sufficient variation in scores
2. standardise date sources to 0-100 scale with mean around 45 and SD around 20.
3. average scores across sources (minimum 3)
4. report standard error and confidence interval.

According to the Mongolia country page, the sources of information for Mongolia are:

1. Control of Corruption: WorldBank (2010)
2. Global Corruption Barometer: TI (2010)
3. Open Budget Index: International Budget Partnership (2010)
4. Global Competitiveness Index: IMD (2011-12)
5. Judicial Independence: World Economic Forum (2011-12)
6. Human Development Index: UNDP (2011)
7. Rule of Law: WorldBank (2010)
8. Press Freedom Index: Reporters Without Borders (2011-12)
9. Voice & Accountability: WorldBank (2010)

Sources that are not available for Mongolia:

1. OECD Anti-Bribery Convention [only OECD countries?]
2. Bribe Payers Index [only “world’s wealthiest and most  economically influential countries”]
3. Financial Secrecy Index

I am a little puzzled by the listing of sources in that none of the more recent scores (I can’t imagine that 2010 sources are driving the 2012 leap for Mongolia) are particularly positive. Since my conclusion now is that the CPI does not seem to involve any judgements by TI independent of the scores from sources, I’m not sure how this score for Mongolia comes about when I look at the components.

Ranking Mongolia

The biggest news is clearly that Mongolia has jumped from 120th to 94th least corrupt country or, alternatively from 62nd to 80th most corrupt.

TI explicitly warns that the methodology for the 2012 version has changed making year-over-year comparison impossible, though the change was in part motivated by the desire to make such comparisons possible in the future.

Mongolia received a score of 36 out of 100 (100 representing no perception of corruption at all). This compares to an average of just under 44 for all 174 countries ranked. The number of sources used in the calculation of Mongolia compares well with OECD and many other countries.

Among the 28 countries included in the Asia Pacific group, Mongolia ranks right in the middle at 14.

Among all 174 countries, Mongolia has an identical score as Benin, Colombia, Djibouti, Greece, India, Moldova, and Senegal. I know very little about most of these countries, but an identical ranking to an EU member country (Greece) and a gigantic Asian democracy (India) is surely not something to be ashamed of.

If we look at post state-socialist countries in the listing, Mongolia sits right in the middle, below European post-Soviet and Eastern European countries (the Baltics, Poland, Hungary, Czech Republic, Slovakia, Romania, Macedonia, Bosnia and Herzegovina, Bulgaria, Montenegro, Moldova), but ahead of other Asian and some European countries (Armenia, Kosovo, Albania, Belarus, Vietnam, Kazakhstan, Russia, Azerbaijan, Ukraine, Kyrgyzstan, Cambodia, Tajikistan, Laos, Turkmenistan, Uzbekistan. Perhaps most notable is the significantly better score for Mongolia than any of the Central Asian countries even though some of the starting point for development in 1990 may have been comparable (though minus the Soviet Republic status, and plus democracy for Mongolia).

I was also trying to group the TI-ranked countries by those that are significant mining jurisdictions, but found no single listing of such jurisdictions on-line. The challenge would be to come up with a metric that would include a country like Canada with a long-established and large mineral sector, and also Mongolia where the volume is small, but the potential is huge.

In any case, eyeballing some countries with prominent mineral sectors (not oil and gas) would put established producers like Canada, Australia, US, Chile ranked highly, but also Botswana, Namibia, Brazil, South Africa. Mongolia would then be somewhat similar among mining countries to Peru, Mexico and the Philippines, but much less corrupt than Indonesia, Kazakhstan, Russia, Nigeria and Venezuela.

Explaining Mongolia

Let’s assume for the moment (and I am happy to make this assumption having looked at the documentation) that the CPI produces reliable information and that the ranking of countries relative to one another is meaningful. This assumption begs the question of what happened in Mongolia in the past 12 months that would lead to such a significant jump. Fortunately, I had a chance to discuss this with PoliSci PhD student Mendee this morning to get a better sense of likely developments.

Remembering that the CPI is a Corruption Perception Index the best explanation for Mongolia’s jump would be to focus on the momentum (Mendee’s suggestion) that has been building around anti-corruption activities in Mongolia. Corruption has been very much in the news this year with the most internationally visible instances being the arrest and trial of former president Enkhbayar and the current detention of Australian lawyer Sarah Armstrong. But there has been much more than this going on that has been much more visible domestically with investigations of several aimag governors, a tightening of income reporting requirements for public and elected officials, etc.

Public perception of anti-corruption activities given a high level of underlying corruption (this is assumed to be the case for Mongolia where people (Mongolian and non-Mongolian) speak of corruption as “endemic”) could be either, “OMG, what a cesspool, corruption is everywhere” (perception as more corrupt than expected), or “Yes, corruption is everywhere, but things are getting better”. Clearly, the second view must have predominated in the sources used by TI.

What makes this explanation of building momentum plausible is that corruption and related topics like conflict-of-interest have become better defined over the past several years in Mongolia. There has been a fair bit of legislative activity that makes some of the principles behind corruption as well as specific definition clearer to all involved so that discussions now are less about the principle than about specifics.

These legal definitions and some initiatives at enforcement may then be acting in two ways, a) by building awareness, and b) by providing a deterrent to potentially corrupt officials.

News Events that may have contributed to perception of more stringent anti-corruption efforts

Obviously, news events relating to corruption have a great potential to change perceptions. When I exchanged tweets with Enkhbold Z, chairman of the Mongolian parliament, he agreed that “yes, I was expecting some improvement. I think creation of ACA [Anti-Corruption Agency], Elbegdorj election [for president 2009], of course Enkhbayar case contributed to this”. While their impact would be most likely on 2013 or even 2014 surveys, some of the 2012 events might have been:

• Arrest and trial of former president Enkhbayar. Contrary to the expectations of Doug Schoen and other participants in an international media campaign apparently orchestrated by Enkhbayar, this did not spell the end of Mongolian democracy, but instead signaled the seriousness of anti-corruption efforts. The Anti-Corruption Agency was very visible in this process, though also accused of being political motivated.
• Tightening and tighter enforcement of election regulations around the June parliamentary election. This included the prosecution of some winning candidates for electoral fraud.
• New Minister of Justice Temujin has obviously gained some prominence in terms of judicial reforms which is in part an anti-corruption effort as well.
• While investigations of mining companies, including South Gobi are primarily viewed as being motivated by political/corruption revenge or “resource nationalism” by foreign investors, these may be signalling powerfully to a domestic audience that anti-corruption efforts are being bolstered.

Caveat: A Secular Trend in Rankings of Mongolia

One alternative explanation to this positive sense of building momentum would be that the information used by TI has simply gotten better. This is not implausible in that more attention is being paid to Mongolia from many directions so that more information is becoming available. There may thus be a secular trend in part associated with the status of having the highest GDP growth in 2011 that would lead to a rise in these kind of index scores for Mongolia.

Conclusions

It has been over a year since the Luis Vuitton boutique was established in Ulaanbaatar. It appears to be economically viable. This alone suggests rampant corruption in a country with very few domestic industrial activities (as of yet) and a significant proportion of the country living in poverty. Perhaps this should be an indicator used by TI for their Corruption Perception Index, but maybe not as it would paint as negative a picture of the country as the ownership of the only Rolls Royce by a minister would. The picture suggested by the CPI is more positive.

The bottom line thus could be, corruption is rampant in Mongolia, but anti-corruption efforts have been stepped up significantly, so a decline of corruption can be reasonably expected in coming years.

Follow