Monday, August 10, 2015

Poverty, Corruption and Economic Growth

One of the puzzling things I’ve found in the empirical data is that corruption doesn’t seem to have any impact on economic growth (I found instead that corruption primarily impacts the variance of growth, but not growth itself or the level of development). It appears to matter what type of corruption is involved, and the institutional framework that a country has.

Here’s Ricardo Hausmann on the same subject (exceprt):

Fighting Corruption Won’t End Poverty

CAMBRIDGE – Countries are poor because governments are corrupt. And, unless they ensure that public resources are not stolen, and that public power is not used for private gain, they will remain poor, right?

It certainly is tempting to believe so. Here, after all, is a narrative that neatly aligns the promise of prosperity with the struggle against injustice. As Pope Francis said on his recent trip to Latin America: “corruption is the moth, the gangrene of a people.” The corrupt deserve to be “tied to a rock and cast into the sea.”

Perhaps they do. But that won’t necessarily make their countries more prosperous.

Consider the data. Probably the best measure of corruption is the World Bank’s Control of Corruption Indicator, which has been published since 1996 for over 180 countries. The CCI shows that while rich countries tend to be less corrupt than poor ones, countries that are relatively less corrupt, for their level of development, such as Ghana, Costa Rica, or Denmark, do not grow any faster than others.

Nor do countries that improve in their CCI score, such as Zambia, Macedonia, Uruguay, or New Zealand, grow faster. By contrast, the World Bank’s Government Effectiveness Indicator suggests that countries that, given their income level, have relatively effective governments or improve their performance, do tend to grow faster.

For some reason – probably related to the nature of what NYU’s Jonathan Haidt calls our “righteous minds” – our moral sentiments are strongly related to feelings of empathy in the face of harm and unfairness. It is easier to mobilize against injustice than for justice. We are more enthusiastic to fight the bad – say, hunger and poverty – than to fight for, say, the kind of growth and development that makes food and sustainable livelihoods plentiful.

Sometimes switching from the “bad” to the corresponding “good” is simply a matter of semantics: to fight against racism is to fight for nondiscrimination. But, in the case of corruption, which is a bad that is caused by the absence of a good, attacking the bad is very different from creating the good.

This largely jives with my own exploration of the data (the article goes on to talk about the importance of institutions). Corruption is worth fighting against for its own sake, but don’t expect success here would suddenly breed prosperity. Of course, in the kind of climate we’re in right now, I doubt anyone’s really listening.


  1. You might be interested in Lant Pritchett's take on this issue:

    He and his colleague tease apart the de jure and the de facto in measuring the effect of corruption/red tape on ease of doing business:

    "Given these two well-established facts about developing countries—a de jure legal environment that creates complex and burdensome regulatory procedures, together with weak governance in implementation—the question arises: how is business actually done? In India, compliance with the formal rules to get a construction permit would take 25 procedures and 186 days. On the other hand, one could get a driver’s license in Delhi without taking the legally required driving exam (or knowing how to drive) by hiring an agent to facilitate the process. In Cambodia, the Doing Business data reports it would take 20 procedures and 652 days to get a construction permit, but in 2012, its “government effectiveness” was −.8 and “control of corruption” −1.0. If government enforcement is not very effective and bribery is fairly common, how long does it really take?"

    1. Thanks John, I've seen it.

      I think the general take on this is that it says more about the (ir)relevance of the Ease of Doing Business Rankings.

  2. Luckily I have not neutered my Reason to cuckold my Conscience…..yep that’s about right. For without Reason as thy lantern in the wreathing darkness andf ading light whither but lost doest thou wander my neutered Conscience……ah...the perils of Inferno lie await ahead unbeknownst to thy Desires………

    Hahahaha….thought a little Shakespeare speak might lighten things up….

    Corruption and growth… Jekyll and Hyde all suited up together…..hahahaaha.

    Unfortunately despite the lame attempt by the writer in the link above to airbrush the reality of a corrupted wasteland. Readers can approach the topic by accessing it here:


    (read the Exec Summary if nothing else but better still the whole stuff)

    or you can read the cancerous impact of Corruption such these:

    To get an idea of how large these effects are, imagine, for example, that India were to ‘adopt’ the (perceived) corruption level of the United Kingdom. As a consequence of this, India’s growth rate of genuine wealth per capita would go up from about 1 per cent to 3.4 per cent per annum. Or alternatively, one can ask by how much (perceived) corruption in the United Kingdom could increase before the UK economy became unsustainable? Using the OLS estimate from regression 23 and keeping initial GDP per capita and geography fixed, the answer is that the TI index would have to fall from 8.65 to just below 2. Countries with
    such a low score include Russia and Uganda. Using, instead, the IV estimate from regression 24, the UK economy would become unsustainable once its TI index fell to the level of Italy (which has a score of 4.7). All in all, the estimates point to economic as well as statistical significance.

    which in plain speak means corruption has a substantial impact on the growth rate of wealth per capita …..

    and further down the paper on sustainable economic development in the long run. All available here:

    (full paper downloadable by the way, after gating issues are resolved)

    So you see folks, Corruption, like Inequality, is a malignant cancer to socioeconomic well-being. No doubt there is no direct causal link between Corruption and Growth established as yet due to all those intervening variables making such a task difficult, one on one relationships as established in the OECD document (as in 1 above) clearly shows the net negative impact of corruption so much so the World Bank was moved to say this:

    ………significant enough to have induced the World Bank (2011b) to declare corruption as “...among the greatest obstacle to economic and social development” by undermining the rule of law and weakening the institutional foundations on which sustainable development depends.

    Cheers...Stay safe everyone and good night Malaysia wherever you are.....

    Warrior 231

    1. @Warrior

      Dude, that paper covers almost the same grounds I did (up to and including the positive correlation seen in East Asia).

      The big difference is that I don't accept the negative correlation between GDP per capita and corruption indicators as anything of significance. By the very nature of the data, the correlation between GDPC and corruption is spurious - there CANNOT be a causal relationship between them.

      You might also want to read this. Even the GDPC/corruption correlation becomes questionable when you look outside the high income economies.

  3. Is this article reliable...

    A typical worker in Selangor saw his income grow by an average RM20 a month. But when petrol prices went up by 10 sen, this translated into roughly a RM32 increase in spending on petrol per month, he said. “So essentially, the rise in income for the average Selangor worker was wiped out by the increase in expenses,” Muhammad told the student leaders forum. - See more at:

    1. @anon 7.21

      I was sitting next to him when he said it, so I suppose you could take that as being reliable

  4. I heard from my brother, who is currently doing business in China, that China's past double-digit GDP growth rates were mainly powered by corruption. Of course, that was his laymen view.

    He said, specific GDP KPIs were dished down along the gov hierarchy, i.e. from central to provinces to municipals to counties to towns to villages. Each 'CEO' will inflates the projects figure of, say, building a bridge for $500m to $700m, in order to hit their KPIs. What makes the country so strong is that the 'CEOs' will still deliver their works in perfection (delivering the full $500m bridge without flaw) while pocketing the extra cash ($200m into pockets). Anyway, $700m is recorded as GDP.

    With Xi Jingping now fighting corruption, China is "struggling" with growth. From a common-sense perspective they are still growing at an impressive rate (given their huge base) but the general opinion (especially from the economists [no offense Hisham]) say that they are doing badly. Whatever it is, corruption seems to be a strong function of GDP growth in general.

    Correct me if I'm wrong. Eager to hear a view from an expert like you Hisham on this matter.


    1. @Fung

      I wouldn't go so far to say that China's growth was "completely" driven by corruption - not everything stems from government spending. The private sector is large and growing. But it's certainly true that the anti-corruption drive has taken a toll on growth, even outside of China. Hong Kong and Macau have both been badly hit (less gamblers and luxury goods purchases), but we feel it even here.

      As for China's numbers, the time series are too regular and smooth to be real.

  5. I am currently reading a book edited by Edmund Terence Gomez titled Political Business in East Asia which is relevant to this. Basically, the book explores the relation between state, party and private businesses in countries like Malaysia, China, Taiwan, South Korea, and Japan.

    There is an important distinction what type of corruption is present in the state-party-business relations. Andrew Webeman makes the distinction between what is called degenerative and developmental corruption. Basically, most east asian countries are practicing developmental corruption where a steady supply of patronage contracts and rents are distributed by state and party officials to private interests in exchange for political support and capital during elections. That is why high levels of corruption are accompanied by high levels of growth in these type of patronage networks. Now this is not asserting that this patronage network fuels economic growth, but rather it has provided a check and balance against the state and party to excessively plunder the economy. Wherein in countries like Haiti and the Dominican Republic, the state engaged in degenerative corruption in plundering the economy excessively which has reduced its growth in the long term.

    My point is that this political alliance with the state, party and private businesses have managed to maintain a steady growth rate while having considerable levels of corruption. That is why there might not be any significant relationship between corruption and economic growth if the state engages in developmental corruption with the objective of spurring economic growth in the private economy through patronage relationships and getting electoral support in return.

    1. @ho

      Thanks, that's spot on, and certainly echoes what I've read. I'll have to look up Gomez' book.

  6. Corruption may be a positive for the economy in the short run eg over inflated projects.But the negative impact will come later with the higher cost of living (tolls,electricity) or subsidies/bailouts (Hiways & LRT).

  7. Corruption may be a positive for the economy in the short run eg over inflated projects.But the negative impact will come later with the higher cost of living (tolls,electricity) or subsidies/bailouts (Hiways & LRT).