22 May 2010

Getting the best out of people: performance incentives for NGOs in #Tanzania

I've been working for NGOs in Tanzania for long enough to know that staff aren't always as hard working and motivated as their bosses (and donors) would like. Many NGO staff get powerful motivation from the work they are doing for the benefit of others, but that's not true of everyone. And even the most dedicated staff have ups and downs, not always working to their full potential.

The private sector has long recognised the usefulness of linking staff salaries and other rewards to performance. Think of mobile phone voucher sales, for example. Sales staff are put on commission. Distribution networks are set up so that each link in the chain sells on to the next at a slightly higher price. And think of the performance-related bonuses and share options paid to senior executives in big business. Even taxi drivers in Dar pay a daily "fee" to the taxi owners, getting rid of the need for trust and/or complicated accounting and giving drivers every reason to maximise their own revenues while protecting the owner's interests. At all levels, it's all about "aligning incentives" so that staff (and other key actors) benefit financially if they and the company perform well.

The contrast with NGOs is stark. What's standard practice in business is almost unheard of in NGOs. Can we not learn something from the private sector and improve our performance as a result?

Of course, it would have its challenges, some of them major. It would need flexibility in budgets, which is harder to manage in non-profits than in the private sector (since a company that makes a bigger profit than expected has more money available to pay out as bonuses). And would donors allow room for such additional uncertainty in programme budgets?

Even if you can solve those problems, what would you use to measure performance? NGOs don't make profit or have a share price. And most NGOs don't have sales either. Many NGO staff have jobs that are varied and unpredictable, which makes setting targets difficult and turns assessments of progress against those targets into a subjective judgement - who would be responsible for making that judgement if someone's salary depends on it?

Even if objective performance measures can be found, they can easily create an incentive to focus on one, more easily measurable, part of a job and neglect the trickier, less predictable, less managable aspects of a job - the very parts that require the most attention, most expertise, and may well be most critical for success.

Ideally, the best performance measures, at least for senior staff, would be the same indicators used to measure progress against the programme's overall objectives. But if we're honest, many NGOs already have major problems with monitoring and evaluation. And it is often programme managers themselves who are responsible for collecting this data.

These might seem like big obstacles, but we should remember how much would be gained from finding a way through them. It has to be worth trying.

And with that in mind, perhaps I could propose a couple of ideas. First, find those individual positions that are well-suited to performance-related pay, and try setting some small bonuses for meeting agreed targets. In some cases, there are objective and measurable indicators in line with programme objectives. Pay researchers extra for finishing a study on time, community mobilisers for meeting a target number of citizens signing up, and even drivers for going 12 months without an accident.

Second, to encourage teamwork and collective responsibility, all the members of a programme team could get a bonus if the programme as a whole meets agreed targets. These would need to be either objective measures and/or judgements made by an independent actor.

In both cases, it would be simpler to start small. Reduce salaries by, say, 10%, but in return, promise that if low range targets are met, they will get the 10% back, and if higher targets are met, they will get an extra 10% on top of their original salary. This could be done twice over for each staff member - once for individual targets and one for team targets. So they get 20% less than their original salary if they fail to meet both their personal targets and the team's objectives, but get 20% more if they exceed both.

It's far from a complete answer, but it's a start, and I'm sure something along these lines could be made to work. It's got to be worth a try. As to how to get donors on board, that will have to be left for another time.