That’s the provocative question raised by Prof. Maxim Mironov of IE Business School in a new paper, which I came across via this post at the Wall Street Journal. In short, Mironov concludes that yes, on occasion, it is worthwhile to hire a corrupt manager based purely on financial performance. Specifically, Mironov finds a positive correlation between smaller firms with corrupt managers and higher revenue growth. As a firm grows, however, the effect disappears (an important distinction WSJ failed to note). More on why this may be the case later.
What I find most interesting about Mironov’s work is its reliance on a large corpus of statistical data in order to make insights about individual behavior. This is not quite ‘big data’, but it is a similar concept and often tends to be perceived as having a more authoritative weight than analyses relying on other sources of information. Of course, the data is only as good as the model you plug it into, a concept that seems lost among the celebrations that liken big data to a contemporary Tower of Babel.
Mironov’s Model – Propensity to Corrupt
At the heart of Mironov’s analysis is what he calls an individual’s Propensity to Corrupt (“PTC”) – essentially, a score indicating how likely it is a given individual will pay a bribe. How does one gather data that represents a statistical sample and breaks down on at the individual level? Mironov’s answer is driving records. Specifically, he first calculates the expected number of recorded traffic violations for each drive based on demographic variables. He then calculates the difference between the expected number of recorded violations and the actual number of recorded violations for a particular driver. The resulting number is that driver’s PTC (Mironov calculated the PTC for over 3 million Muscovites).
As you can see, the underlying assumption of the PTC is that, in aggregate, the most likely explanation for drivers having significantly less recorded traffic violations is that they paid bribes to traffic police to avoid an official fine. Mironov concedes, however, that it may also be due to “other unobservable characteristics related to driving safeness” (i.e., safer drivers have less traffic violations). But presumed corrupt payments are also an “unobservable characteristic”. This is where the data’s reliance on the model is made clear: the PTC is only as strong as the assumption that less violations generally equals more bribes.
Measuring PTC Effect on Firm Performance
Next, Mironov tests whether PTC correlates with firm performance, where he correlates individual driving histories with employment records of the top five highly paid employees at Russian companies (in order to ensure that he is measuring the PTC of the firm’s management, and not rank and file employees). He then looks at those companies’ financial performance, both in terms of reported earnings and bank receipts. Including the latter measure avoids companies that underreport earnings to cheat on taxes, which you might expect from a company with corrupt managers. As mentioned above, he finds a positive correlation between PTC of top managers and a company’s revenue growth – both reported and in terms of bank receipts – for smaller companies.
Mironov also looks at the PTC for members of certain government agencies, and identifies the geographic jurisdiction of these agencies. He then measures the effect that PTC of a given agency has on firm performance in the jurisdiction. Unsurprisingly, he finds that more corrupt government agencies (i.e., higher PTC) are negatively correlated with revenue growth of firms within the agencies’ jurisdiction. In other words, contrary to the “efficient grease” theory of corruption cutting through red tape, bribe-seeking authorities tend to suppress economic growth at the individual firm level.
Explaining the Results
Mironov’s Explanation – Entrepreneurial Self-Selection
To his credit, Mironov’s favored explanation for his findings is both unexpected and novel. He argues that, in a corrupt country such as Russia, creative/talented people with high moral/ethical standards prefer other professions to entrepreneurship. I am not sure I am completely convinced by this analysis, although we do know that “bureaucrat” has become the most desired job among Russians, for obvious reasons. So perhaps business attracts individuals with a higher PTC to begin with, which means businessmen with higher PTC are more motivated to be in their chosen profession, and thus perform better regardless (i.e., correlation of PTC with firm performance does not equal causation).
Predictive Value of Traffic Corruption
Another issue is that Russian traffic police are consistently ranked as one of the most corrupt institutions in Russia. Russian traffic police are also notorious for stopping vehicles for almost non-existent violations in order to hunt for bribes. In this situation, a driver may have a whole host of pressures pushing him towards paying a bribe to avoid a ticket (especially in the case of a shakedown) aside from a desire to avoid a legal impediment.
Thus, it is debatable whether observed traffic corruption has any predictive value with respect to other forms of corruption (e.g., to obtain a building permit, win a contract).
Why Only Small Firms?
More importantly, Mironov’s results only applied to smaller firms, with revenues of USD 1.55 million or less, and particularly applied to firms with revenues between USD 117-217K (see graph).
Why would smaller firms be more likely to reap the benefits of corrupt managers? There are several possible explanations:
- With less revenues to speak of overall, the boost provided by an expedited building permit or electrical connection provide a proportionally greater boost to smaller firms as compared to mid-sized and large firms. In other words, smaller firms benefit from corrupt managers to a greater extent the same way they benefit from charismatic managers to a greater extent – the effects are magnified.
- Smaller firms that have comparatively more interactions with government officials are in a ‘growth’ phase because they are opening new facilities, bidding for new tenders, importing more goods, etc. Thus, higher growth results in more bribe demands, which require more corrupt managers. Here, high growth and corrupt managers have a symbiotic relationship.
- Smaller/medium may also face more corrupt demands overall than their larger peers. Indeed, the most recent data from the World Bank’s Enterprise Surveys indicates that medium enterprises (20-99 employees) are much more likely to be expected to give gifts to secure a government contract, secure an electrical connection, or obtain an operating or import license.
- Larger firms may also be deeper into its corrupt relationship with government officials, and perhaps with a greater number of officials. Rather than the demands ceasing once payment is made, companies pay a continually greater share as they grow until their obligations to corrupt officials create a natural limit to high growth.
There are other considerations that should give companies pause before embracing corrupt managers. The obvious one is that bribery is illegal under Russian law, and most foreign companies have laws in their home countries prohibiting bribery abroad. Also, although bribery is common in Russia, it is not socially acceptable (and in fact has become less so over the past four years). Thus, a manager willing to engage in corrupt behavior is signaling a general openness to violate general ethical/moral principles, including those that protect the company (e.g., theft). A morally ambiguous manager’s behavior will set the standard for employees down the line, and the result in many cases will be a rotten culture that is not easily eradicated.