Talk to Graduate School of Government, University of Sydney
I want to give you a primer on the pros and cons of government intervention in the economy. While, as public servants, you take government policy activity for granted - it’s what you’re employed to do - the appropriate role of government (whether, and under what circumstances, governments should intervene in markets) is perhaps the most contentious topic in politics and economics. Let’s take a quick look at an aspect of the politics of intervention before we take a much closer look at the economics.
The political philosophy of intervention
The political philosophy of libertarianism - which gives primacy to individual liberty and carries a presumption against the need for government intervention - is overrepresented in the political debate in Australia and particularly the US. While by no means all economists are libertarians, most have a big streak of it in them because the dominant model of conventional, ‘neo-classical’ economics is built on three key assumptions that almost inevitably bias it against intervention. Those who come to the neo-classical model from a political perspective (giving primacy to individual freedom) rather than an economic perspective (giving primacy to the best management of the economy) adopt a fundamentalist, no-questions-asked approach to the model.
The first key assumption is that people always act ‘rationally’ in the decisions they make. That is, they act with clear-headed, carefully calculated self-interest. One of the commonest catch-cries of the libertarians is: how could the government possibly know what’s in my best interests better than I know myself?’. Well, if we all were as rational as the model assumes we are, that would be a killer argument. In reality, however, most of us are often far from rational in much of our decision-making. We act on instinct, we’re swayed by our emotions, we don’t pay enough attention - don’t read the label, the instructions or the product statement - and we tend to do what everyone else is doing. In which case, it’s perfectly possible - in principle, at least - that governments could know what’s in our best interests better than we know ourselves.
The second assumption is that markets are self-correcting or self-righting - that, in the jargon of economists, they have an inbuilt tendency to return to equilibrium. You don’t need to study the behaviour of the financial markets to doubt the veracity of that proposition. Sometimes it happens; many times it doesn’t.
The third assumption is that society consists solely of individuals - individual consumers, and firms so small relative to the size of the market they have no ability to influence the market price. So the possibility of people acting collectively - whether voluntarily or by electing a government to make decisions on their behalf - is simply excluded from model. It admits no circumstance where, by co-operating rather than competing with each other, we could achieve a superior outcome.
Put the three assumptions together - we’re always rational, markets are self-righting and individual actions are the only ones available - and you see why the only thing government intervention could do is stuff things up. Hence the advice to governments: laissez faire - leave things alone.
While the rhetoric of libertarians and some economists implies that markets have always existed and government intervention in markets is a much more recent and unwarranted intrusion, this is not historically accurate. Though it’s true humans have exchanged goods (traded with each other) for millennia, markets in the form we know - the market-based economy - are a much more recent development, dating from the dismantling and replacement of the feudal system. Markets are actually the creation of governments because they rest on government creation and enforcement of private property rights. And much of governments’ actions and interventions over the centuries have as their primary or secondary objective enhancing the functioning of the market-based economy. Think of the regulation of money as a medium of exchange and store of value, bankruptcy rules and commercial law. Even government spending on universal education and health has huge spin-off benefits for the market economy. Quite clearly, there never has been or ever will be such a thing as a ‘free market’. All there is are markets in which governments intervene to a greater or lesser extent.
Libertarians set the liberty of the individual as their supreme value, which must never be compromised. All of us set a high store on personal freedom. But most of us accept there are many other worthy objectives of government that are often in conflict with the untrammelled freedom of the individual. So the most sensible approach is to find the best trade-off between freedom and other important objectives - other dimensions of the public interest - being willing to diminish our freedom to the extent the conflicting objective is sufficiently important.
So much for libertarianism. Fortunately, economists take the question of intervention a little bit more seriously and go a lot deeper. Let’s start again, relying on an article that appeared in the 1995 edition of the Asian Development Bank's Asian Development Outlook.
The assumptions of the neo-classical model
Microeconomic theory starts with a simple model of markets in which there is 'perfect competition'. It says that, in the absence of government intervention, the interaction of self-interested consumers with profit-maximising firms will produce the most efficient allocation of the economy's resources. That is, those resources will be used to produce the particular combination of goods and services that offers the maximum satisfaction of consumers' material wants.
But to reach this desirable conclusion, the model relies on a host of assumptions. Most elementary textbooks list four key assumptions: the market must consist of large numbers of buyers and sellers; every firm must be selling an identical ('homogeneous') product; all buyers and sellers must have complete knowledge of all relevant prices, quantities, conditions and technologies; and there should be no barriers that prevent firms entering or leaving the market.
To these better-known assumptions, however, the Asian bank article adds four more: there should be no spillover or external effects, so that all parties bear the full costs and receive the full benefits of their production and consumption activities; there should be no unexploited economies of scale; all parties must know their own best interests; and there should be no uncertainties or ambiguities.
Do those assumptions strike you as realistic? Can you think of a market in which all of them hold true? Of course not. As the Asian bank says, 'these assumptions are extreme and unrealistic in their literal form'. And that's why this idealised model of perfect competition is merely the starting point of the economists’ theory of markets. 'Despite these glowing theoretical results’, the article continues, 'real-world markets may well be deficient in one or more of the necessary assumptions of the theoretical model and thus may fail to deliver the ideal efficiency that the perfect-competition model promises.'
Causes of ‘market failure’
The next step in the theory is to identify the circumstances in which markets will fail to deliver the goods. The bank lists at least seven kinds of 'market failure'.
First, market power. If there is only one (monopoly) or a few (oligopoly) dominant sellers in a market, and if entry by new firms isn't easy, the established sellers are likely to exercise market power. That is, prices will be higher and quantities produced will be lower than those promised by the competitive model. As well, quality may be lower, varieties limited and innovation diminished.
Second, ‘externality’ effects. If the actions of producers or consumers affect people outside the market, then the market's outcomes are unlikely to represent an efficient allocation of resources. In cases where these external effects are unfavourable - the generation of air or water pollution, for instance - the uncorrected market outcome yields too much of the particular activity, with prices that are too low and with too little effort made to reduce the unfavourable spillovers. In cases where the external effects are favourable - such as an innovation or new idea that others can use - the market outcome yields too little of the activity, with prices that are too high and with too little effort made to increase the externality.
Third, public goods. The two key qualities of public goods are that they are ‘non-rivalrous’ (my consumption of the good doesn’t reduce the quantity of it available to others eg knowledge, use of the internet) and ‘non-excludable’ (no one can be effectively excluded from using the good eg free-to-air television). The standard examples of public goods are lighthouses and defence spending, but there are other, less perfect examples. The free market will produce less of a public good than is in the best interests of the community because it’s so hard for private firms to make sufficient profit from producing it. This is why governments often end up producing those goods and services which have partial or complete public goods characteristics. In practice, most of the services governments provide - including health care, education, law and order, defence and much more - are thought of as public goods.
Fourth, economies of scale. If firms aren't producing in high enough volume to exploit economies of scale fully, then their activities won't achieve allocative efficiency.
Fifth, incomplete information and uncertainty. If sellers and buyers don't have compete information about how products work, the alternative products, the range of prices and even about future events, their production and consumption decisions won't yield efficient outcomes.
Sixth, asymmetric information. If, as is often the case, the sellers know a lot more about the product and the market than the buyers do, then market outcomes will not be efficient.
Seventh, the 'second best' problem. If there are uncorrected market failures in one market, then perfect competition in related markets is unlikely to yield efficient outcomes even in those markets. That's because all markets are interrelated. It follows that, if a distortion in one market can't be corrected directly, a second-best solution may be to induce compensating distortions in related markets.
The economists’ ground rules for intervention
It's because economic theory identifies all these potential forms of market failure that intervention in markets is commonplace. But while the public is always urging governments to intervene to correct problems, real or perceived, and politicians are almost always keen to leap in, economists have a two-stage test before they accept such a need: First, a significant instance of market failure has to be demonstrated and, second, the ability of government intervention to correct the market failure - or at least do more good than harm - has to be demonstrated.
Causes of ‘government failure’
This brings us to a more recent development in economists’ theory of markets, which focuses on the possibility of 'government failure'. Government failure arises where government intervention to correct market failure worsens outcomes rather than improving them, or where the modest benefits don’t justify the considerable costs (eg the various subsidy schemes for household solar power). If we're going to talk about real-world markets, we also have to talk about real-world governments. The Asian bank’s article lists at least four sources of government failure.
First, ill-defined goals. Governments often have very broad, ill-defined and even conflicting goals for interventions. In practice, trying to achieve conflicting goals can lead to arbitrary and inefficient outcomes.
Second, weak incentives and poor management. With ill-defined goals and the absence of a profit motive, public employees are likely to face weak incentives for good performance. Good management is a scarce skill and is usually highly paid. Where top salaries aren't high enough, governments find it hard to attract and retain high-quality managers, thus worsening outcomes.
Third, information problems. Governments may encounter as much or almost as much difficulty in acquiring full information as market participants do.
Fourth, 'rent-seeking' behaviour. Specific interest groups will seek to use the forces of government to create special favours for themselves at the expense of others in the community. For instance, special subsidies, tax breaks or limits on competition. They invariably seek to justify this behaviour by claiming that it's in the national interest or even that it would correct market failure.
The theory of ‘public choice’
This brings us to the relatively recent political/economic theory known as ‘public choice’, developed by James Buchanan and Gordon Tullock. The theory holds, among other things, that politicians and bureaucrats always act in their own interest rather than the public’s interest, and that, whatever its original motivations, all government regulation of industry ends up being ‘captured’ by the industry and turned to the industry’s advantage in, say, reducing competition within the industry (to the incumbents’ advantage), increasing protection or in persuading the government to subsidise industry costs. The regulated have a huge incentive to get to the regulators so as to modify the regulation in ways the industry finds more congenial, or to advantage the existing players against new entrants or rival industries.
I don’t accept for a moment the accusation that all regulation of industry is subverted. But I do believe there’s more than a grain of truth to the accusation: there is considerable scope for regulatory capture. And I’ve often suspected that the way most bureaucracies are organised - where the department of agriculture looks after the farmers, the industry department looks after the manufacturers, the environment department looks after the greenies, the resources and energy department looks after the miners and the tourism department looks after the tourist industry - could have been purpose-built for regulatory capture. In the various industries’ battle for their share of industry assistance, in the inter-departmental battle for influence and resources, each industry has its own special champion, those whose true role is supposed to be to keep the industry acting within the bounds of the wider public interest. Is the bureaucracy divided up this way just to gain the benefits of specialisation, or is each department’s real role to keep their particular industry happy and not making trouble for the elected government?
Another dimension of potential government failure arises because governments - and government departments and agencies - have some of the characteristics of a monopoly. They deliver public services funded by the taxpayer and there are no alternative suppliers. Monopolies are almost always bad, becoming lazy, unresponsive, self-serving and high-handed in their treatment of the individual members of the public they are supposed to serve, who can be seen as ignorant inconveniences. It’s enormously tempting to deliver services according to rules than suit the department rather than the ‘client’.
I’m never greatly impressed by all the libertarian rhetoric about ‘the nanny state’. But they do have a good point. Governments simply can’t solve all the problems we face in our lives, so we do need to be wary of weakening the ordinary person’s acceptance that the first responsibility for solving their problems rests with themselves. We’ll be helping people who can’t help themselves, and in certain circumstances we’ll be providing universal assistance but, for the most part, it’s down to you. It’s too easy for talkback radio to expect a government solution to every problem that comes along, an expectation that’s fed by the way politicians on both sides seem to be promising just that in every election campaign.
Then there’s the related problem economists refer to as ‘moral hazard’: the more people know they’re covered against risks, the less hard they try to avoid those risks, thus leading to excessive claims for assistance. This is problem with all forms of insurance, which insurance companies try to counter by such devices as no-claim bonuses and high co-payments (‘front-end deductibles’).
My conclusions from the debate
Where I do stand in this debate? I believe market failure is common and that governments should usually act to correct it. But I also believe in government failure and some degree of truth in the public choice critique. Governments and their bureaucrats do sometimes act in their own interests rather than the public’s and some regulation is captured and perverted by those being regulated. So I believe in intervention, but I’ve been around long enough to know it’s a very tricky business, with enormous potential for creating perverse incentives and other unintended consequences. We need to work hard to get the intervention right, minimising unintended consequences and doing more good than harm. This requires a lot of careful thought, trial and error, experimentation, learning from experience and project evaluation. This is why I’m pleased to see you studying Policy in Practice and interested in discussing the choice of appropriate policy instruments.
Some general principles for improving intervention
First, avoid ideological extremes. Because the truth is a hard-to-find position somewhere in the middle, it’s tempting to seek the simple certainty of one extreme or the other. But the sensible position is to be neither opposed to almost all intervention nor indiscriminate in intervening. The hard part of bureaucratic endeavour is to find the sweet spot, where interventions do more good than harm. Avoid prejudiced assumptions that the private sector is always more efficient than the public sector, or that the public sector is always more committed to quality than the private sector. Take a pragmatic, evidence-based approach.
Second, rationalise policy objectives. A great advantage of the private sector is that everything it does has a single, simple objective: to make money. For governments, things are never that simple. They have, and should have, multiple objectives. But while it may be possible to kill two birds with the one stone, it’s never possible to kill five. Politicians always what to use the same dollar to satisfy a host of interest groups and almost always lack the discipline to set priorities among all the things it would be nice to do. But multiple objectives are usually conflicting and unless bureaucrats can reduce that conflict the chances of interventions being ineffective are high.
Third, respect the power of market forces. To deny the infallibility of market forces should not be to underestimate their power. Self-interest is a hugely powerful motivator, not just among the public but also within government and the bureaucracy. And people do change their behaviour in response to changes in prices - sometimes irrationally so. When you try to suppress market forces they usually pop up somewhere else, like squeezing a balloon. People will look for and exploit the inevitable loopholes in your regulations; if there’s a system they’ll game it.
Fourth, by the same token, remember the limitations of the conventional model. Those limitations are so pervasive it’s not surprising interventions lead to so many ‘unintended consequences’. People aren’t rational; they’re influenced by their emotions, by perceptions of fairness and by what everyone else is doing. The model ignores all incentives apart from monetary incentives and disincentives, whereas non-monetary incentives - motivations, would be a better word - are often pervasive. For instance, people can work hard because they’re ambitious for power and promotion independent of the extra salary, because they love what they’re doing, because of a work ethic or a sense of duty, because of the institution’s esprit de corps. Sometimes the creation of monetary incentives - paying people to do things - can be counterproductive if it crowds out pre-existing non-monetary motivations. SES performance bonuses may be a case in point.
Fifth, try to work with the grain. Market forces are so powerful it’s often better to harness them in the service of the regulatory objective than try simply to stomp on them. This is the rational for the ‘economic instruments’, such as trading schemes and pollution taxes, used in environmental regulation. Even so, environmental subsidy schemes can often be terribly wasteful, and in specific areas the best approach can be direct intervention - legislating to raise motor vehicle emission standards or to require the weatherproofing of new-built homes.
Sixth and finally, remember the gold standard of intervention: voluntary compliance. The best laws are laws that don’t need to be enforced because so many people comply with them voluntarily. Why would they? Because they’re actually conforming to the norms of socially acceptable behaviour. Humans are social animals, preoccupied by the desire to fit in, meet the approval of their peers, be like everyone else and be no more antisocial than others. Interventions that undermine existing social norms can be far more unsuccessful and damaging than expected.
Many interventions - whether direct rules about what people may or may not do, or numerical or monetary incentives, such as KPIs - can be so onerous in robbing people of autonomy and ability to exercise their professional judgment that they become counterproductive. People stop trying and caring, and switch to looking for loopholes and ways to cheat the performance measurements. Much better to find ways to get people to internalise the values of the institution or the society, so they do what’s wanted of them out of a sense of duty, loyalty and just the satisfaction of knowing they’ve made their contribution and performed their role well. Intrinsic motivation always trumps extrinsic motivation.
Remember, however, that well-judged interventions, which use the force of law to change people’s behaviour in socially desirable directions, can end up being reinforced by the development of new norms of acceptable behaviour. Why? Because, contrary to everything rationalists assume about how the world works, people seek to reduce their cognitive dissonance by changing their values and beliefs to fit their behaviour. Force me to change my behaviour and I’ll change my values to fit. Changed attitudes towards sexual harassment in the workplace, smoking indoors, and drinking and driving are among the many examples of this process in operation.
Read more >>
