In the book “When Crime Pays”, author Milan Vaishnav makes the argument that crime and muscle reaps rewards in Indian politics. Vaishnav states that the level of crime increases proportionally with the candidates’ income levels -- less than two percent of the candidates with the lowest income faced criminal indictments, as opposed to over nine percent of the richest ones. Corruption and crime dilute with time as a democracy matures. Till then, there is need to give each voter more bang for her buck -- the rewards of voting needs to increase dramatically.
We are a country where the last mile delivery of services is a challenge and the day-to-day is a struggle for most. If the elected local representatives are more engaged and responsive, it can make a marked improvement in the quality of life for the common folk. One can argue that motivation of re-election and a sense of duty maybe incentive enough, however there is merit in the idea of adding a layer of financial incentive to the mix.
It is not politically correct to discuss financial incentives for politicians. People are both cynical about our political class and yet expect them to be selfless do-gooders and hold them to a higher standard (as they should).
As reputed economist Tim Harford puts it, it is extremely challenging to execute even the most noble of intentions in a corrupt society, which inevitably breeds apathy. Strategically placed financial incentives can stir the whole pot into action and aiding the development process. Taking a pragmatic approach to things as they are (and not how they should be) may help tilt the needle towards the desired outcome and allow space for such an uncomfortable idea.
Outside the political arena, performance based financial incentives are widely employed. For example, in financial investing the origin of the incentive fee structure can be traced to Alfred Winslow Jones and Warren Buffet who in the late 1940s / early 50s established the “2/20” model. This was inspired by Phoenician merchants, who charged a fifth of profits from a successful journey. Fund managers raise a corpus from investors. Two percent of the total is used as management fees and 20 percent of the profit is distributed amongst the managers. The same exists across the private sector -- in healthcare as well, studies have shown that financial incentives lead to an overall enhancement of performance during surgery, especially amongst those with low baseline performance.
The question of measuring performance in matters of governance is trickier. The World Bank, for example, divides governance activities into programmes, which have macro level ‘goals’. These programmes are further divided into micro-level, time-bound targets, each of which is ascribed an agent who is accountable, thereby easing the process of monitoring and evaluation. Researchers at the University of Southern California have provided a list of specific indicators. For example, in the case of a road construction programme, the indicators can be divided into inputs (money, personnel, equipment), outputs (miles of road built), productivity ( number of workers used), costs (average cost per mile), public satisfaction (surveys on quality of roads). Within each indicator, it would be possible to determine desired levels, and offer a financial incentive upon attainment of those levels on time.
In India, the development funds allocated to various types of elected representatives could prove for an interesting pilot / testing ground.
For example, Members of Parliament Local Area Development Scheme (MPLADS) is a 1993 scheme that allows each MP to suggest projects to the tune of INR 5 crore per annum to the district head, to be taken up in his/ her constituency. The objective of the scheme is to enable MPs to recommend developmental work for creation of “durable community assets” such as health and education. Given the relatively manageable size of the budget, and the measurable nature of the programs, the MPLADS scheme would serve as an interesting experiment for key performance indicators based (KPI) financial incentive structures. One could also add non-financial incentives such as giving recognition awards (public accolade) to the better performers.
Policy findings from the developing world show that financial incentives in the public sector can work if they are measurable, simple and linked to clear targets. In the education sector, a 2011 study from Andhra Pradesh found that in a large scale relatively small incentives of 3% of teacher’s annual salaries significantly improved student learning in math and language. In another example in India incentives improved teacher attendance and in turn student learning. In public finance, an IGC study found that offering incentive payments to tax collectors in Pakistan substantially increased tax revenues (increasing collection by 13% in certain areas). Research on health centres in Rwanda suggests that performance-based pay to health staff might lead to improved pre- and post-natal care. Similar experiences have been found in Indonesia.
Colombia has been successful in its monitoring and evaluation of government programmes through its SYNERGIA tool, using the programme-based structure of the World Bank described above. In 1969, Malaysia introduced the Program Performance Budgeting System, which essentially involves outcome-based budgeting with the aim of achieving globally competitive levels of development. The system was reformed to put even greater emphasis on outcomes in 1990. Similarly in Africa, both Uganda and Egypt have taken considerable steps to improve monitoring and evaluation systems.
Research has extended this argument particularly to politicians, and the evidence is encouraging. Scholars have found that higher wages correlate with increased representation of minorities. Others have found a positive correlation between salaries and the quantity of legislation passed. It has also shown to increase the quality and behaviour of the political class. Scholars have also argued that the incentive of re-election can induce politicians to address less relevant issues - commercial levers can tilt the balance towards harder problems. Evidence from Brazil’s municipal legislature the Cˆamara de Vereadores suggests that higher wages result in more educated candidates, increase in length of stay in office, and the rolling out of key development projects.
There have been similar discussions in the developed world as well. A 1991 editorial from the St. Louis Post-Dispatch proposed an amendment to the US Constitution, linking reduction in national debt with annual bonuses for members of the House. Similarly, an editorial in 1997 in the Tampa Tribune argued for a link between raises for legislators and rate of growth of household income. Martin Skladany in the Yale Human Rights and Development Journal (2014) also argues towards a scheme of performance based incentive bonuses for politicians. From the US, evidence also suggests that higher wages can lead to better congruence between citizens and Governors. In Europe, political parties have used financial incentive to have Gender Balance. Parts of Canada have also tried the same. Beyond politics, Bush Administration used financial incentive to encourage public schools to better their performance.
Such ideas are based on the “efficiency wages theory” championed by Akerloff, Stiglitz and Janet Yellen, which demonstrate that pay matters. Others have also studied the same under the banner of “incentive based contracts”. There is also research on the psychology of incentives.
Such an idea is not without its limitations. First, in general what works for the private sector will not always work in the public sector. Second, there is concern that incentives might “crowd out” intrinsic motivation to get a job done. However, there is little empirical evidence to support this. Instead, new research about Zambia and Nigeria find that incentives tend to leverage intrinsic motivation. Third, performance incentives can backfire when outcomes are vague or difficult to measure. These lessons are hard learnt from trials with Kenyan teachers and US police officers. Fourth, there is also fear of gaming the system as was learnt by a study linking financial rewards to nurses’ attendance in India.
Incentives should be based on either input or output depending on the context and what is easier to measure and harder to manipulate. A much broader lesson is that each incentive has to pilot tested in a sandboxed environment keeping in mind the specific environment. There are few studies around non-financial incentives as well. In a public health example from Zambia, agents who received social recognition for their sales achieved double the sales vs. those earning a commission of up to 90% on condom sales. Another example was found amongst tax collectors of Pakistan. Promising results were found in a performance-based postings scheme that gave officers an opportunity to be transferred to more desirable locations depending on revenue collection performance. Out of the box non-financial incentives can raise performance without un-due burden on the exchequer.
Further, in the example of Indian politicians, the proposal has certain specific version limitations. First, in the context of MPs, the elected official only has advisory power with the actual execution in the hands of state and district authorities. Second, there remains the open question of financing the proposal -- reducing from the existing the development fund is not advisable. Third, publicly announcing the payout and thus the “performance” of each MP may not be politically practical and a decision in private maybe untenable for other reasons. Fourth, measuring KPIs in social audits is not always a perfect science and subject to errors and manipulation.
An Indian MP gets an all inclusive monthly amount of INR 2,00,000 (the government spends on average INR 2.7 lakhs per month per MP). There have been debates in Parliament around pays. The current Lok Sabha is the richest in the Indian history with 82 percent members having disclosed assets worth INR 1 crore or more. One can argue that against these numbers, financial incentives will not move the needle. However there is no harm in experimenting with a clearly defined measurable scheme.
New suggestions in politics can be controversial, as the former Indian chief economic advisor Kaushik Basu learnt when he proposed to legalise some kinds of bribes. The above analysis is purely hypothetical, of course, and there are limitations to its implementation in the Indian context.
Nevertheless, it is important to encourage such hypotheticals and foster debate, particularly around issues that provide such a substantial challenge to India’s development. As citizens we have a certain duality in our nature. We assume the worst amongst our politicians and yet expect the best from them. This constant friction prevents the spread of ideas that match reality. This proposal maybe unpleasant to some because it assumes the worst amongst our politicians. However, as elucidated by Vaishnav, the self-interested behavior of a politician should be aligned with the broader interests of the country.
Some of these limitations are more serious than others. However, the point of new ideas is to stir a debate and propose a broad framework. Moreover changing the behaviour of even a few actors on the ground will have a meaningful impact. Steps will be needed to be taken before translating such an idea into action. For example, what kind of financial and non-financial incentives and the quantum of financial ones. Measurable KPIs will need to be defined and an audit board be established within the Ministry of Statistics and Programme Implementation. It may also compel an eventual reform in the projects sanction procedure.
In order to test the above hypothesis, the first step would be a multi stakeholder consultation. Next would be to identify a single issue (e.g. public health) in a limited jurisdiction (e.g. limited constituencies) and take a sandbox-like 'pilot' approach before scaling up.
Strong evidence is needed before deployment of a policy at scale (understanding results, cost-effectiveness, unintended consequences and other learnings). With learnings, it is feasible to extend the same framework to the State, the municipality and the panchayat. In essence, while there are things beyond money that bring value to public office, it is nonetheless a factor to be considered. Since 1995, India has had an average ranking of 75 (from 175 countries) on the Transparency International Corruption Index. In order to improve these numbers and to meet our global aspirations, there is merit in thinking of new ideas, irrespective of how novel and counter-intuitive they may be.
It is true that politicians must be driven by a higher cause and lot of them are. Yet the right kind of measurable target based incentives (financial and non-financial) could add a lot of desired flavour. At the end, people vote for performance. Can we create incentives where one corruption does not compromise performance and two it allows entry of better people at least at the margin?
The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.