It is well known that there has been something of a “performance movement” (as Beryl Radin has called it) in western countries in recent years. (I have written an account of UK developments for the World Bank myself here).
There are many reasons cited for this movement, but the one that usually catches the interest of the political classes is that they can demonstrate greater accountability and show their ability to deliver by performance monitoring and improvement and thereby enhance their credibility and (say it quietly) their electability.
But what if that is not true, or at least not in any simple way?
There is some evidence – and I don’t want to exaggerate this, it is only suggestive at this stage – that there is a more complex relationship between performance and electoral pay-offs in western countries than one might instinctively expect.
Research conducted on English local government is very suggestive (Greasly and John, 2010). Studying the relationship between the performance of local governments and their subsequent electoral success (or lack of) researchers found a curious asymmetry in the relationship – bad performance led to political penalties (incumbents tended to loose elections) but good performance did not lead to political pay-offs – above a certain level performance seemed to be discounted by electors.
This result reminded me of the research on ‘happiness’ levels in society which suggests there is a ‘plateau’ in economic advancement where above a certain level, further gains in wealth do not produce increases in happiness, which tends to level-off (see The Spirit Level by Wilkinson and Pickett).
And it prompts the obvious question – is there a parallel here with citizen’s satisfaction with the performance of public services?
This may be a case where a Western innovation – massively increased performance reporting on public services – is actually likely to have greater political impact in emerging economies than it does in the west. In the hypothesis that there is a sort of ‘hygiene factor’ at work and that gains in public service performance can have big payoffs in developing economies, it is potentially important lever for change which political elites could easily be convinced to buy into.
Of course, the converse is true for the west – whatever other good reasons there might be for the performance movement – and I for one think there such good reasons (see my latest book ‘Theories of Performance’) – political advantage would clearly be a fairly weak one.
Reference
Stephen Greasley, Peter John. (2010) “Does Stronger Political Leadership have a Performance Payoff? Citizen Satisfaction in the Institutional Redesign of Sub-central Governments in England.” Journal of Public Administration Research and Theory
STATECRAFT - improving government 
Very interesting post, Colin. Wilkison and Pickett’s main theme is around the impact of inequality in that equation, but as you mention, there is an additional sub-theme regarding the relationship between performance and other measures, which seems to be not relevant in *developed* countries beyond a certain threshold (ie when it is hard to offer substantive, visible improvements to the public). Once you reach life expectancy above +77 in a country, improvements in healthcare services are usually pretty marginal.
This argument offers a very pessimistic conclusion for developed countries, indeed. However, I think that this argument is less true in periods of technological revolutions (eg 1996-2000), when citizens’ and politicians’ expectations about what it is feasible are completely redrawn and high, and everybody gets quickly excited around any tiny improvement related to that upcoming technological revolution. That momentum was very handy for the Clintons, Blairs, Schroeders and Aznars of the time… But this “momentum” never can last long, as everybody eventually realizes that the most substantive problems of the system are not solved (just marginally impacted) by technological change.
We have been doing research at the World Bank trying to measure if/when good performance and service delivery become a driver of political support and trust in government institutions.
One of the pieces that were discussed last Thursday measured the impact of Service Delivery improvements (in basic infrastructure, health, education and crime) on the levels of political and institutional support, looking at the experience of Medellin, Colombia (2003-2009). The research was drawing on previous theoretical work on the potential causal drivers of trust in government (Manning et alt. 2010, Manning and Wetzel 2010), and on the hypotheses that were raised by Medellin public officials, business groups, and civil society organizations during preliminary interviews. The qualitative analysis suggested that a quick upgrading in the city’s less favored districts, combined with a stronger enforcement of the rule of law, may have successfully raised the traditionally low levels of political and institutional support in the city, breaking the equilibrium of distrust in government. This specific research provided additional quantitative empirical evidence (microdata) supporting the insights gathered in the qualitative and theoretical work, and it also assessed the relative importance of perceived procedural (or administrative) fairness, perceptions of improvements in service delivery, actual service delivery, and satisfaction with services, as possible alternative channels to rebuild trust in institutions. Taking into account spatial (across households and neighborhoods) and inter-temporal (across years) dimensions in Service Delivery, this research found that *perceived* improvements in different public services may have a distinct impact on citizens’ perceptions of government performance and trustworthiness. In addition, it seems that there are potential spillover effects of trust across institutions, suggesting that citizens can actually distinguish between performance of politicians and institutions, especially if these institutions/government agencies are very VISIBLE and relevant for citizens’ wellbeing (salient).
The two main insights from this emerging research are that the impact of service delivery improvements on political support/trust in government is larger when: (i) visibility of implementing public agencies and institutions is very high (data showed that it varies a lot across institutions, eg. the police or the local tax office to file your taxes vs what’s happening inside the planning ministry or the civil service office), and (ii) these improvements take place in areas that citizens prioritize *today* (as assessed by surveys on citizens’ worries and priorities). A third element would be related to your posting: when the level of service provision is really high already, the margin for delivering significant improvements in that sector is narrow, and as citizens tend to reward improvements in service delivery from last year which are *above the expectations*, then it is hard to obtain
The moral of this emerging evidence is that there is not a straightforward relationship between improvements in service delivery and political/trust pay-offs: politicians/civil servants need to be strategic and choose key areas (ie in priority sectors, and increasing the visibility of the reformed institutions as much as they can) so they can maximize the “political return to investment”.
Sorry for the long posting.
What if happiness/satisfaction is not a reality, but simply a social construction? No doubt we all feel happy sometimes, but we also become angry, dissappointed, bored, stressed, relaxed, fascinated, interested, concerned, drunk, etc., etc. Everyday life is filled with feelings. Maybe it is simply an illusion that there is an average happiness in a individual as well as in a population.
Let’s suppose that the overall satisfaction is a social construction. Next question is: How and by whom is it construed then? It seems that we have a number of agents that construe happiness/satisfaction by selecting feelings and events and interpreting them: Media, politicians, spin-doctors, administrations, etc. And not to forget: People themselves. Thus, the satisfaction of administrative performance depends on a complex reconstruction of the performance itself and how it is experienced – as well as a lot manipulation. Could it be otherwise?