Good Meritocracy, Bad MeritocracyFeb 01, 2013 at 11:40 am | Hits: 30633
By Donald Low
Meritocracy is widely regarded as a core principle of governance in Singapore. Basic disagreement over it was also one of the reasons why Singapore separated from Malaysia. The meritocracy principle – that we try to equalise opportunities not outcomes, and that we allocate rewards on the basis of an individual’s merit or his abilities and achievements – is as close as anything gets to being a national ideology. In recent years however, there has been growing disquiet over meritocracy as it is practised in Singapore: the excessive competition it engenders (particularly in our education system), the stress and anxiety it causes, and the inequality that many see as the result of an unfettered meritocracy.
As other commentators have pointed out, the choice facing Singapore is not a binary one between meritocracy and other ways of distributing rewards. Rather, the critique is over how meritocracy is practised in Singapore. How we translate the meritocratic ideal into practice matters more than the fact that we subscribe to the principle of meritocracy in the abstract.
To help frame the debate, I would like to propose three questions which I hope can inform the meritocracy debate in Singapore. In particular, I want to argue that there are varieties of meritocracy – that some forms of meritocracy are desirable, while others are possibly malignant. The debate in the National Conversation should not be over whether we embrace meritocracy or not; rather, it should be over the kind of meritocracy we want.
Defining Meritocracy: Rewarding Type vs Rewarding Effort
The first question that should guide our reflections on meritocracy is whether it should reward type or effort. The main justification for a system that rewards people by merit is that it encourages effort. An economist might argue that meritocracy helps society deal with the problem of moral hazard that is likely to arise with other ways of allocating rewards. When rewards are tied to one’s abilities and achievements, people are motivated to strive and be the best they can be. In contrast, when rewards are tied to inherited wealth, connections or race, people’s behaviours might change in morally hazardous ways: corruption and rent-seeking behaviours increase, people channel more resources to socially unproductive activities that benefit themselves (and their families) but not society.
But it appears to me that the meritocracy practised in Singapore – especially in our education system – is one which rewards type more so than effort. That is, our meritocratic system seems to reward people who possess the “right” attributes. An economist might describe our meritocracy as one which aims to solve the problem of adverse selection – the problem of sorting or distinguishing the good from the rest. Adverse selection problems are, for instance, common in insurance markets where the insurer often cannot distinguish between good risks (those he wants to insure) and the bad risks (those he does not want to insure). Consequently, the insurer devises ways to get his potential customers to signal or reveal their riskiness; he also chooses to insure only the right risk types.
Like moral hazard, the problem of adverse selection is also a problem of information asymmetry. Both problems arise because one party is unable to perfectly monitor or observe the behaviour or quality of another party. But whereas the moral hazard perspective highlights how people’s behaviours might change when the rules (for allocating rewards) are changed, the adverse selection perspective focusses attention on how quality (or merit) is determined.
A meritocracy which is focussed on rewarding type as opposed to rewarding effort suffers from at least three disadvantages. The first is that we may define or measure merit poorly, yet assign too much importance to these imperfect measures of merit. Many commentators have pointed out how the PSLE system – a very precise way of dealing with the adverse selection or sorting problem – is too consequential in that a single examination at a relatively young age determines too much of a person’s subsequent education outcomes. Given reasonable doubt over whether it is capturing and assessing merit correctly, the great importance that the current education system attaches to PSLE scores is clearly problematic.
Second, when a meritocracy is overly focussed on sorting the best from the rest – with more rewards channelled to the former – the incentive for individuals to gain a competitive edge through non-meritocratic means is accentuated. Starting positions, parental background and connections begin to matter more; inequality in this generation is more likely to be reproduced in the next. In the context of our education system, the preferential access to schools that the children of alumni enjoy, the higher concentration of good schools in affluent neighbourhoods, and the greater resources that well-to-do families have for tuition and enrichment programmes are examples of how our meritocracy might contribute to inequality and lower mobility.
Third, a meritocratic system focussed on rewarding type forces people to compete for relative position or advantage. A system that measures people on relative as opposed to absolute performance is more likely to be wasteful. The analogy that is commonly given is that of everyone at a stadium standing to get a better view. While individually rational, this is wasteful for the group as a whole. When everyone decides (rationally) to put in more resources for the positional good (a better view), relative positions are unchanged but everyone “spends more” just to maintain their relative position. We would all be better off if we can come to some binding agreement to spend less in this “arms race”.
This kind of collective action problems characterises many parts of our meritocratic system. For instance, it ails our education system where students are graded on a curve in examinations. It also explains why parents probably spend much more on private tuition than what is collectively rational. Competing for limited places in choice schools – for example, through parent volunteer schemes – is yet another example of a zero-sum arms race in our education system.
Another problem with a meritocratic system which allocates rewards based on relative performance is that it reduces trust and cooperation. Anyone who has ever been in contention for a promotion in which only one person can get the position will understand this. Since my success now depends on my outperforming the others, I have no incentive to help my colleagues succeed. Meritocracy shapes individual behaviours and social norms. When the form of meritocracy that is practised rations rewards by relative performance, it could promote a selfish, “me first” mentality and erode norms of trust and cooperation.
A meritocracy based on relative performance may also not raise average performance even as it “stretches out” the group’s performance and raises inequality. One might argue that when rewards are allocated on the basis of relative performance, the incentives for individuals to excel are stronger. Average performance should increase. While this is probably true for top performers, the effects on the rest of the group are less predictable. For instance, those further down the ability curve might feel discouraged and opt out of the system that they feel they do not have a reasonable chance of success in. Stress and (performance) anxiety are also likely to increase. As epidemiologists have shown, higher stress levels produce cortisol in our bodies; over sustained periods, this increases the risks of illnesses such as hypertension, psychiatric disorders and cancers. As in an arms race, everyone loses out when they compete for relative position.
The argument here is not that we should allocate rewards based only on absolute performance and avoid measuring ourselves on relative performance in all contexts. That would be unrealistic. Sometimes we have no choice but to compete for relative position. Some things are intrinsically scarce, and competition for these things will inevitably be zero-sum. One thinks of the university valedictorian, the Olympic champion, limited university places, the CEO appointment, and many other contexts where we have to accept that scarce rewards are – and should be – allocated on relative merit.
But when we compete for relative gain, we should recognise that this form of meritocracy imposes potential costs on society. This makes it incumbent on the government (which has the main responsibility for dealing with collective action problems) to find ways to ameliorate the harmful effects of competitive “arms races” – especially through fiscal redistribution. I will return to this point on redistribution later in the essay.
As a society, we also have a degree of autonomy to decide on the things we want to use an absolute yardstick to determine rewards, and the things we decide on the basis of relative merit. In some contexts, we can choose whether we want to be a meritocracy that rewards effort or one that rewards type. For instance, while it may seem natural to us that admission to secondary schools should be based on relative merit or type, the fact is that in many other countries with high-performing education systems, examinations (graded on a curve) are not used to determine high school admissions. Even when such examinations are used, they are seldom used to the point where they are as consequential as ours are.
Regulating Meritocracy: Wall Street versus Silicon Valley
The second question that should guide our thinking on meritocracy is what rules should govern and constrain the behaviour of those who have done well in the meritocratic system. There is no prima facie reason to believe that those who have succeeded in a meritocracy will channel their energies to socially useful activities. Neither should there be a presumption that our legal and regulatory systems are always able to deter, anticipate and punish the abuses and wrong-doings of the successful. The risks of regulatory capture, and of regulators lacking the information and incentives to do the right thing are all real – even for advanced economies with well-developed institutions.
The global financial crisis which began in Wall Street is a case in point of how an ostensibly meritocratic system can wreak havoc on the rest of society. Wall Street meritocracy is toxic for at least three reasons.
The first is moral hazard. Bankers took on enormous risks knowing that their institutions enjoyed implicit guarantees from the state and would be bailed out if their bets went bad. They enjoyed an unlimited upside even as their downside was capped by the state. This system of flawed incentives, combined with the opportunities created by financial innovation and securitisation, explained why Wall Street banks peddled so many subprime mortgages to people who were not credit-worthy, why they sliced and diced these mortgages into securities which were then sold to the investors who could not easily monitor their quality, and why they paid themselves large fees for increasing risks in the financial system. When these securities lost their value in the face of collapse in the housing boom, Wall Street banks were shielded from the consequences of their bad decisions because these institutions were deemed “too big to fail”.
Second, it is highly doubtful that all the financial “innovation” on Wall Street was socially useful. While securitisation is potentially a good thing (structured properly, it spreads and diversifies risks), it actually increases systemic fragility when the risks of the individual securities are highly correlated and when the maintenance of their values depends on flawed assumptions (like the assumption that house prices would never fall). The former Chairman of the US Fed, Paul Volcker, went as far as to say that “the only socially useful banking innovation (in recent decades) was the invention of the ATM.”
Contrast the meritocracy of Wall Street – where losses are socialised and gains privatised, and where “innovation” resulted in products that are socially harmful – with the meritocracy of Silicon Valley, and it becomes clear that we have two kinds of meritocracy with contrasting outcomes. Nobody is bailed out in Silicon Valley; an entrepreneur succeeds or fails based on whether consumers find his inventions useful. While Silicon Valley is also prone to speculative booms and busts, the damage it causes society when its bubble bursts is far more limited.
Third, Wall Street meritocracy breeds a pernicious, self-justifying, entitlement narrative. For example, Wall Street bankers justified the decision to pay themselves millions in bonuses from the bailout monies they received from taxpayers on the grounds that not doing so would result in talent leaving the financial industry. The kind of meritocracy practised in Wall Street breeds a belief among its beneficiaries that they are entitled to their rewards, that the system is inherently just, and that inequality is a natural consequence of an efficient and normatively desirable system. They view those who have not succeeded in the system as slothful or lacking in talent or merit – and undeserving of state support. It therefore increases the rich’s resistance to the redistributive policies needed to tackle rising inequality. Over time, such a meritocratic system entrenches inequality and immobility; the risks of this society becoming permanently more stratified and divided by class are also much greater.
Needless to say, the kind of meritocracy practised in Silicon Valley is superior to the meritocracy laced with corporate socialism practised in Wall Street. The former is just and legitimate in the eyes of average citizens because the successful bear the full consequences of their decisions and are not insulated from the risks they take. People are more accepting of inequality and uneven rewards when there isn’t one set of rules for the winners and another for the rest. Wall Street suffers from a severe legitimacy crisis not only because the bankers caused so much damage, but also because they are widely perceived to have gotten away with it, and because of their sense of entitlement.
In Singapore, we are frequently reminded of the risks and moral hazard problems of providing too much help for the poor. This is the main justification for why we must avoid a welfare state. The crisis is a powerful reminder that the risks of moral hazard are far greater when the rich and successful are not properly regulated and reined in. Corporate malfeasance imposes a much larger cost on society than the fecklessness of the poor addicted to government welfare. For a longer exposition on the social costs and risks imposed by the rise of the super-rich, Joe Stiglitz’s May 2011 article in Vanity Fair is a must-read.
Another important lesson from the crisis is that not only do markets sometimes fail, but the rules and institutions designed to anticipate and correct these market failures can also fail. In a healthy democracy (and meritocracy), the watchman also needs to be watched. This calls for eternal vigilance on the part of citizens and civil society. It means creating strong safeguards against regulatory capture, increasing transparency and public accountability, ensuring the (political) independence of public institutions, and maintaining a healthy distance between regulatory agencies and the corporate entities they regulate.
Legitimising Meritocracy: Trickle-down versus Trickle-up
The third question is how we reconcile meritocracy with inequality. While we like meritocracy for rewarding hard work and recognising people’s talent and abilities, we dislike the unequal outcomes it produces. Is this an inevitable trade-off? If we want an efficient system that creates strong incentives for people to strive, must we also accept the highly unequal distribution of incomes and wealth that such a system produces?
One way to think about this question is to contrast two worldviews. The first, which might be termed “trickle-down meritocracy”, sees the growth of the economy and the progress of society as driven by its elite, by its best and brightest. A society organised along the lines of this worldview channels a larger share of resources and opportunities to its high performers and talents. A trickle-down system is not concerned with equality of outcomes, but with ensuring that its talents have the room to achieve and excel, and are not shackled by high taxation and excessive regulations.
In such a society, economic efficiency takes precedence over distributional concerns or considerations of social equity. Indeed, this view contends that the poor are best served by providing the best and brightest with maximum opportunities to succeed as they are ones who create jobs for the rest. Holding the talented back by having onerous taxes or regulatory restraints on markets undermines growth and hurts the poor – the very people whom the advocates of social justice claim to help. A trickle-down approach also means that government and society should be concerned more with growing the pie (through market-friendly and pro-talent policies), and less with how the pie is distributed.
In the last twenty years, Singapore society has probably become more of such a society. Income tax rates have been slashed, wealth taxes reduced (for example, the estate duty was abolished in a place that already does not tax capital gains), and public spending as a share of GDP reduced. The state has also not become more redistributive in the face of rising inequality. Consequently, even after taking into account taxes and transfers, inequality today is higher than it was a decade ago before government redistribution.
A second worldview, which one might term “trickle-up meritocracy”, sees government redistribution and fair outcomes as necessary corollaries to market-friendly, pro-capital policies and the meritocratic system. According to this view, meritocracy is legitimate only if it benefits the bulk of society – not just in absolute terms, but also in relative terms. This is because greater inequality reduces subjective well-being, social mobility, and trust. Meritocracy with high and rising inequality sows the seeds of its own demise. If we care about preserving meritocracy, it is incumbent on us to limit the rise in inequality.
Trickle-up meritocracy also differs from the trickle-down variant in at least two other ways. First, it contends that ensuring equal opportunities is necessary, but not sufficient. Because people start with differences in talent and resources, equalising resources at the start to some extent is justified on the grounds that this is necessary to ensure equal access to opportunities. Such a system is still meritocratic as the equalisation is done at the start of the competitive race and does not diminish incentives for everyone to run as fast as they can. The race is competitive and meritocratic, but the state has intervened to adjust starting positions and given those with fewer resources a head start.
Second, trickle-up meritocracy believes that instead of subsiding and extending tax breaks to the rich in the hope that they will create jobs and prosperity for the rest, fiscal policy should be focussed on increasing the human capital of the rest, ensuring they can afford basic needs like housing and good healthcare, assuring them of retirement security and social protection against contingencies like involuntary unemployment, and reducing inequality. In the long run, these measures also benefit corporations and those further up the income ladder.
Spending on these social goods, which is likely to require vigorous fiscal redistribution in the context of today’s globalisation and rapid technological change, is necessary to save meritocracy from itself. Indeed, without these redistributive measures, meritocracy rests on increasingly shaky and tenuous foundations.
In the national debate on meritocracy, let us avoid framing the issue as a false choice: meritocracy or no meritocracy. A constructive and meaningful debate would instead focus on whether ours is a meritocracy that rewards effort or type; what rules and institutions we should have to regulate those who have succeeded in the meritocratic system; and what kind of social spending would maximise “trickle-up” benefits.
I wish to thank David Skilling of Landfall Strategy Consulting for his critique of this essay. His distinction between a meritocracy that rewards effort versus one that rewards type was extremely useful in helping me organise and frame the ideas in the essay. I also benefited from his reflections on the Australian and New Zealand education systems.
Damien Huang, a research associate at the Lee Kuan Yew School of Public Policy, provided valuable insights and suggestions when I was developing this essay. I am grateful for his help in sharpening the ideas in this essay.
Donald Low is Senior Fellow and Assistant Dean (Research Centres) at the Lee Kuan Yew School of Public Policy, National University of Singapore.