Incentives and Development Conferences

A few days ago, I started studying for the GRE.  Needless to say, my dormant procrastination skills have awoken with rancour.  Today these skills are being put to blog use. 

Policy makers and development programme designers seem to love incentives.  Want women to deliver babies at accredited health centres?  Give them a cash reward for doing so. Need health counsellors to enrol more people for tuberculosis traetment? Try cash incentives.  Malawians don’t return to clinics for their HIV test results?  Give them some $.

Everyone loves a good incentive (source:

Humans do respond to incentives. So why not use them on ourselves?  More specifically, I’m thinking about international development conferences.

A few months ago, I went to a development related conference in Colombo.  It was hosted at a beautiful hotel on the coast and attended by people from all over the world.  A bilateral aid agency paid for my tickets and conference fees, and gave me a very large corpus of cash for other expenses (accommodation, local travel, dinner, alcohol, shopping, laundry, etc).  Without being in any way thrifty, I saved a few hundred dollars.  I did not need to submit any receipts and was allowed to keep the excess money.  I wonder if having an incentive to spend less could reduce the overall costs of such events.  Perhaps reimbursment based on actual expenses and a cash bonus to the person who spends the least?  The amount of the bonus would need to be precisely calculated, and there would have to be a mechanism to ensure that people don’t just say “I spent $0”.  So the bonus would have to be less than the actual expected expense, else everyone would just say “I spent 0”.  There would also need to be a cap on expenditure.  The admin costs of this system may negate the benefits, but I would be interested in seeing some experimental data – randomly assign some conferences to be fully lump summed, some to be have half participants on lump sum and half on incentive scheme, and some to have only the incentive scheme.  This could also give us some interesting insight into how much development professionals value these conferences…

The second similar area where incentives could be used is choosing locations for conferences.  I am not sure how this is presently done, but surely there should be an open bidding process for the larger events.  Recently, the 10th International Conference on Aids in Asia and the Pacific was held in Busan, South Korea.  It is unlikely that Busan, as wonderful as it may be, was the most efficient venue for this event.  Korea is one of the more expensive countries in Asia, and getting to Busan is not nearly as convenient as hubs like Singapore, Bangkok, or Hong Kong.  In addition, Korean is not widely spoken in the region, and I don’t know how many Busanese speak English.  To be sure, there are advantages of hosting in a wealthier city – it’s likely that Busan has more stable water and electricity, as well as access to other facilities, than Ulan Bator or Port Moresy.  But either way, an open bidding process, where potential organizers in different cities can bid for larger conferences, could incentivize more tightly managed international development events.  Big agencies should also publish data on how much these events cost.

Coming soon:  a series on auto economics.  That’s not doing economics on oneself.  It’s the economics of auto rikshaws in Delhi.  Should be fun.

In defence of the NGO SUV

My boss recently told me that the India chief of a very large environmental INGO moved here from a corporate sector job in Singapore and took a paycut of 50%.  His annual income went from INR 4,00,00,000 (that’s four crore to you non-South Asians out there – USD 896,000) to half of that.  One could look at this in two ways:

1. This guy’s awesome – he left $$$ to do some good in the world.

2. Whoa this guy is making so much $$$ while working at a non-profit…how immoral.

Which brings me to the main point of this post – why does conventional wisdom tell us that development type work should not be monetarily well rewarded?

That's our tax money at work

It is easy to think of the development sector as being completely different from the corporate world.  In the latter, it is assumed that people will act in ways to maximize personal and institutional profit.  In the former, personal motivation is supposed to come from some intrinsic desire to “do good”, while institutional performance is measured not by the bottom line but rather by some (often abstract and arbitrary) development indicators.  There has been a lot of interest in the last few years on better defining success, the most visible example being the Millennium Development Goals, but it seems to me like the labour market in development still works on the assumption that people are in this industry to do good.  There are many problems that this creates:

1. It makes hiring and firing processes very complicated. How do you rate someone who is awesome at managing projects but doesn’t care about the poor?  Can you fire someone for being shit at their job but really caring about poverty?

2. Looking at market wage rates, development sector professionals are usually underpaid. While I don’t know if such a study has been done, I would hypothesize that if we take, for example, all the economics graduates from a particular liberal arts college in the last 20 years and look at their income in the first 5 years after graduation, then we will find that those working in NGOs, think tanks etc are paid less than their peers in the corporate world, even when controlling for GPA and such.

3. Given that money flows in development are largely top-down, the idea that one should be paid below market rates follows the money and finds its way to the lowest level workers too.  This is a big problem in the Indian context (and I would imagine others too), because there are a large number of people who are in development because they just need a job and want to support their families and not because they want to save the world.  These guys start out with NGOs, get some training and become more qualified in, e.g., doing surveys, then leave for the private sector where they earn a lot more money.  There is little incentive to stay in development when they see that even their superiors are earning less than entry-level people in, e.g., a for-profit market research firm.

Now to the SUV, which is often used as a symbol of the cost-inefficient ways of aid workers (side note – hilarious SEAWL post on drivers).  The nice car serves a very important function that is easy to ignore – incentive to work hard and stay in the development industry.  This is probably not true when foreign “experts” visit “the field”, since locals are acutely aware that those guys are just special, but rather when local superiors hang out with local field staff.  Status indicators like nice cars (and they’re Toyotas…hardly super fancy) and, much more importantly, competitive pay cheques, will encourage people to continue working in organisations doing socially desirable work.  If donors are serious about wanting to find local solutions etc, then they need to create a system that encourages locals to look for these solutions.  Right now, there is a much greater incentive to join a better paying job in the private sector.

There is obviously much more to this topic but I fear this post is already long enough that, if you’ve made it this far, you’re probably alone. I don’t mean this to be a defence of all aid spending and I have no doubt that current mechanisms are hugely inefficient (e.g. one of my friends at the IMF told me they installed new TVs at the DC headquarters so people could watch the football World Cup – that is ridiculous). But I think the decision makers in this sector are woefully out of touch with field realities.  While World Bank senior economists may have given up lucrative jobs in the private sector to work in development, the majority of local level development workers are probably more interested in the pay cheque than in the social benefits of their work.  Local solutions are great, but there needs to be a structure which incentivizes finding them, and SUVs might be part of that structure.