Talk to Federal Court judges’ continuing education day, Brisbane, March 19, 2015
It’s normal for economists in my position to give you a happy chat about the state of the economy - or the world economy - full of forecasts about how things will turn out and warnings about what the economic managers need to do to ensure everything turns out well. But we all know many of my predictions would turn out to be wrong, and those that were right would probably be right for reasons I hadn’t foreseen. It’s one of the great mysteries of modern life that economists can be so bad at forecasting without the public ever tiring of asking them for more.
But such talks are usually accompanied by a PowerPoint presentation with copious graphs, and I’ve never got the hang of PowerPoint. In any case, though I have three years of economics in my commerce degree, I don’t regard myself as an economist, but rather as a journalist who writes about economics. I’m not a member of the economists’ union. This allows me to act as a kind of interpreter and go-between, standing between the economists and the public. I see my role as providing my readers with a critique of economics and economists, much as our theatre critics provide our readers with a critique of the latest plays. My goal is to explain and demystify economics, advising my readers on when they ought to accept the advice of economists and when they shouldn’t.
So today I want to talk about economics and economists, with special reference to that mystifying animal, economic modelling. The world is full of experts but, particularly in the area of public policy, few experts’ advice gets taken as seriously as economists’ advice. I imagine the economic aspects of the matters that come before you aren’t often kept in the background; that you often have people mounting economic arguments - or arguments that seem economic to the unqualified mind - and that some of you even have economists appearing before you as expert witnesses.
I find it interesting to speculate about why economics and the advice of economists get taken so much more seriously these days than they were in years past. My theory is that it’s because we live in an age of heightened materialism. There’s nothing new about materialism, of course, we’re all materialistic to a greater or lesser extent, and those who never spare a thought for the material aspect of their lives will be lucky if they don’t starve. But I think we’ve become more materialistic than we were - you can see that in the way the aspirations of entrants to American colleges have changed over the years - and this leaves us as a community more inclined to listen to and act on the advice of economists. But, I suspect, it’s a two-process: the more we listen to economists, the more materialist our attitudes become. So, if I’m right, the dominance of economic advice is both an effect and a cause of our era of heightened materialism.
Economics is the study of how market economies organise the production and consumption of goods and services. In other words, it deals with a very important aspect of life - all of us are consumers and most of us are producers - but only one aspect. It’s preoccupied with the practical, material aspect of life.
But there are lots of ways to define economics and another way is to say it’s the study of ‘the economic problem’, which is the problem of scarcity. Scarcity arises because our resources - of land, labour and capital - are finite, whereas our wants are infinite. Scarcity in this context doesn’t mean as scarce as hen’s teeth, just that things aren’t free - they can be acquired only at a price. So economists advise the community on ways to use our finite resources to satisfy as many of our wants as possible. This explains economists’ preoccupation with efficiency: the more efficiently we use our resources, the more wants we can satisfy, the more bang we can get for our buck.
The conventional economist’s first piece of advice to the community is that the most efficient - or least inefficient - way for a community to organise production and consumption is via markets. Governments will need to ‘hold the ring’ in a market economy, conferring and enforcing private property rights, enforcing the law of contract, providing bankruptcy laws, providing companies with limited liability, imposing adequate standards of financial reporting and auditing, and the like. There will be other instances of significant ‘market failure’ but, for the most part, most conventional economists believe that government regulation of markets, or intervention in markets, should be kept as limited as possible so the ‘invisible hand’ of market forces can play its semi-miraculous role of causing all things to work together for good.
If that sounds sarcastic, it’s not intended to be. There are two mistakes you can make about market forces: to underestimate their power or to conclude they’re near infallible. I’m about to bring to your attention the limitations and weaknesses of conventional economics and economists and their advice, but I’d hate to leave you with the impression I regard economics as little more than a hoax. I certainly don’t. The world would be much poorer without the contribution of economists and their discipline. They may not be much chop at forecasting, but that doesn’t mean they don’t know more about the workings of markets and the macro economy than the rest of us. If had more time I’d be happy to give you a list of their most valuable insights, but I think it’s more important for you to understand their limitations.
I want to make five points about economics and the economists’ art, before making five points about economic modelling.
The first point about economics is that it’s narrower than many people realise. We assume that economists are experts on the economy, but it’s truer to say they’re experts on the markets that help to make up the economy. Their basic model - known as the neo-classical model - is a model of how markets work through the interaction of supply on the one hand and demand on the other. This interaction determines the price of the good or service in question and it’s the market price that that brings demand and supply into balance (‘equilibrium’). This means that economists are obsessed by prices. If economists wore tee-shirts they’d say: Prices make the world go round. Economists believe the key to the efficient allocation of resources - to making our finite resources satisfy as many of our infinite wants as possible - is to get the price right which, for the most part, you do by leaving market forces free to determine that price.
Second, like all professions, economists suffer from what I call ‘model blindness’. Just like model trains or model planes, economic models consciously simplify complex reality. They’d be of no use if they didn’t. The idea is to include and highlight the key factors and get rid of the unimportant issues that merely cloud the workings, thereby capturing the essence of what causes what. The neo-classical model strips away other commercial considerations so it can get to what economists regard as the heart of the matter, price. The question to ask of a model is not whether it’s left things out, but whether what it’s left out is important. And the test of that is how good it is at predicting how people (‘economic agents’) will behave in given circumstances. I believe that, in many circumstances, the standard model’s prediction record is poor. That is, it leaves out a lot of factors that do turn out to be important.
To put it another way, the motivations for human behaviour are simply too complex to be adequate captured by any model. It’s fashionable to say that economics is ‘the study of incentives’, but that’s just a sexy way of saying economics is the study of prices. To an economist, prices and the changes in them are the one great incentive - to produce more or buy less, or to produce less or buy more, depending on whether the price has risen or fallen and on whether you’re a producer or a consumer. One major limitation of economic analysis is that it can’t take account of any factor that can’t be not just quantified, but also valued in dollars. Sometimes when a relevant factor doesn’t have a market price because it’s not traded in a market, economists will try to estimate a ‘shadow’ price for it but, for the most part, factors than can’t be monetised are simply ignored. So economists end up tacitly assuming people are motivated solely by monetary considerations. No one does anything out of the goodness of their heart, because of a sense of duty, because they simply enjoy doing a good job or because they’re seeking power and influence. Little wonder economists aren’t much good at predicting people’s behaviour.
The economists’ model has many other limitations but, as with all professions, economists tend to lose sight of the factors that have been excluded from their model. Just as lawyers tend to view every problem from a legal perspective and doctors see all problems as medical, so economists suffer from their own ‘model blindness’ - a tendency to view the world and to analyse problems exclusively through the prism of their model; to focus on those variables their model focuses on and to ignore all those factors from which their model abstracts.
It’s the job of their hearers - including judges - to quiz them about which factors they haven’t taken into account and what their relevance might be. On that score, it’s important to be aware that most economic estimates of the extra income (economic growth) that this or that action would bring about take no account of the ‘distribution’ of that extra income between high, middle and low-income families. In other words, one of the factors from which the model abstracts is questions of fairness. Although a few economists specialise in studying distributional questions, most economists’ analyses ignore it. When pressed, they’ll tell you perceptions of fairness are subjective and so outside their field of competence. So they leave them for others to judge - politicians, for instance. Their specialty is efficiency, not ‘equity’. Trouble is, if you don’t press them they probably won’t mention this limitation to their advice. By contrast, the public tends to regard questions of fairness - who wins, who loses - as highly relevant to the decision-making process.
Third, though it’s fashionable to bang on about the need for decisions to be evidence based, economics is more faith-based than evidence-based. Actually it’s better to say economic analysis is theory-based, with only a secondary appeal to empirical evidence. Whereas the hard sciences collect empirical evidence then try to think of theories that best explain that evidence, economics works the other way. It starts with theories (models) based on assumptions - assumptions which, in the basic neo-classical model, have changed little since the second half of the 19th century - then looks for empirical evidence that supports the theory.
This is not to imply there’s been little advance in the economics profession’s thinking since the 19th century. There has been. Most of the Nobel prizes in economics awarded in recent decades have gone to economists exploring the limitations of the basic model’s assumptions - advances such as information economics and behavioural economics. But most of these advances have been too hard to express mathematically, or too hard to measure, for them to have made much impact on either ‘the economic way of thinking’ or the formal mathematical models used to make forecasts about the economy or about the effects of a particular policy change.
One important but only implicit assumption built into economists’ thinking and their formal models is that the parties to a transaction - say, Woolworths and one of its suppliers, Coles and one of its customers, or any employer and one of its workers - are of roughly equal bargaining power. The inappropriateness of such an assumption does much to explain the government intervention embodied in many of the acts you deal with, including the Competition and Consumer Act.
Fourth, economists and, more particularly, economic rationalists, tend to be missionary or imperialistic in their attitudes. Notwithstanding the narrowness of their model - its focus on the material dimension of our lives and abstraction from the relational, social, cultural and spiritual dimensions; its abstraction from considerations of fairness and significantly unequal bargaining power - economists of a more fundamentalist disposition want us to think like economists think and make economic growth and efficiency in the allocation of resources our overwhelming priority.
Significant parts of our community life have not been part of any market. We don’t allow people to buy and sell blood or body parts, for instance, much of our sport is amateur, many speeches are still given for no recompense other than a bottle of wine and there is much volunteering. But many economic rationalists are so convinced of the benefits of increasing efficiency, and so enamoured of markets as a way of allocating resources efficiently, that they want to see more and more aspects of our lives given over to the market. They want to commercialise and privatise government-owned businesses, and contract out the provision of government services such as job search assistance and childcare. They want to raise more revenue from user charges and less from general taxation. The de facto privatisation of our universities has been proceeding for several decades. Rationalists want to complete the commercialisation of the weekend by completing the deregulation of shopping hours and removing penalty rates. What effect would this have on family life? It’s literally of no consideration; it’s not in the model.
Economic advice is often so missionary, while being so one-dimensional, that it ought to come with a product warning: here’s a list of all the factors I haven’t taken into account in proffering you this advice; you should probably consult a social psychologist or sociologist or cleric before acting on my recommendations. But it rarely comes with any qualification: I’m an expert and here’s what you should do.
Finally, economists have no professional association - no law society or bar council - and so have no expressed or enforced standards of ethical behaviour. The Economic Society is essentially a discussion group that the unqualified are welcome to join. So these days the many firms of economic consultants or the economists undertaking similar work for the big accounting firms, such as KPMG and Deloitte, are under no external ethical constraint when they provide supposedly independent economic advice to paying customers advocating or opposing government policies, or when they appear as expert witnesses.
Which brings us neatly to economic modelling exercises. I make no profession of a deep understanding econometrics, but I’ve quizzed modellers about their models for so long that their limitations are clear enough to me to allow me to make five points.
First, economic models are highly sophisticated and quite primitive at the same time. They’re sophisticated in that they’re a mathematical representation of the linkages between selected elements of the economy. The links are represented by a large number of algebraic equations with numerous variables. The modeller decides the values of the independent (or ‘exogenous’) variables and the model calculates the values of the dependent (‘endogenous’) variables. So if the modeller slots in the wrong independent variables - they’ll usually be assumptions (guesses) about the future - the dependent variables will be wrong, too.
Models are primitive because they’re a hugely oversimplified version of the economy, which often can’t adequately represent the subtleties of the policy changes whose effects they’re purporting to estimate. This lack of subtlety, or a lack of empirical data, means they often make extensive use of ‘proxy’ indicators, a euphemism for ‘the nearest we could find’.
Second, like economic theory itself, economic models are built on a host of assumptions. Indeed, many of those assumptions come from the modeller’s preferred theory about how the economy works. It’s common for ‘computable general equilibrium’ models to have Keynesian assumptions in the short run (up to 10 years) but neo-classical assumptions in the long run (20 years or more). That is, key variables such as inflation, unemployment and economic growth are determined by the strength of aggregate (total) demand in the short run, but by the strength of aggregate supply in the long run.
This means the economy is assumed to be at full employment in the long run, and economic growth over the period is assumed to be determined solely by the growth in the labour force and the rate of improvement in the productivity of labour. Both of those are independent variables, plugged in by the modeller on the basis of some assumption.
So the results you get from such models are largely predetermined by the assumed structure of the economy and the chosen values of the independent variables. Worse, it would be an incompetent modeller who couldn’t tweak the assumptions to ensure the results they got were consistent with what their paying customer was hoping for.
Third, most commercial modellers keep the workings of their models largely hidden. They’ll have some impenetrable qualifications and explanations up the back of their report, but they don’t go out of their way to warn lay users about their model’s assumptions and limitations. Even university economics courses don’t labour the limitations of the theoretical model, and practitioners become so familiar with the way their econometric models work that they can forget about their limitations until they’re challenged. Then, when the results of the modelling are being quoted by the interest group that paid for it - or by politicians on the same side - any remaining restraint is lost and the results are invested with about the same authority and certainty as the Ten Commandments.
Fourth, disinterested parties to whom modelling results are presented should arm themselves with caution and scepticism. All modelling results are rough estimates. Estimates that are expressed too precisely exhibit ‘spurious accuracy’ and are a sign the modeller may be more interest in impressing the punters than enlightening them. Modellers should be asked to outline the key assumptions that are driving the models results and then asked for ‘sensitivity analysis’ of the key variables. That is, if this variable was increased or decreased by 1 percentage point, what difference would it make to the results? Don’t assume that if one economist says an effect is big and another says it’s small, the truth is probably somewhere in the middle.
Finally, be particularly wary of claims about jobs. Businesses are motivated by profits, not a desire to provide employment to people. Economists are on about achieving faster growth in GDP and GDP per person - a crude measure of our improving material standard of living. They know that where there’s growth in GDP there’ll also be growth in employment, but increasing employment is not their primary concern.
But both business people and economists know that when you’re trying to persuade the community to let you undertake a project that will cause economic disruption and environmental damage, there’s just one, killer argument: all the jobs you’ll create. The public have it deeply ingrained in their brain that there’s a permanent excess of unemployed workers and an eternal shortage of jobs, so that you can never have enough jobs. So promise lots of jobs and no further questions will be asked.
This presents businesses and their modellers with an enormous temptation to exaggerate estimates of the number of jobs the project will ‘create’. They’ll focus on the many jobs created in the construction phase, not the much smaller number of permanent jobs. They’ll misuse the multipliers in the Bureau of Statistics’ ‘input-output tables’ to exaggerate the number of jobs to be created ‘indirectly’ in other industries. They’ll forget to remind you that if the jobs to be created are skilled jobs, skilled workers are usually in short supply, so that the jobs won’t be filled by the unemployed but already-employed workers will have to be paid sufficiently highly to attract them away from other employers or from overseas. They won’t mention that if the economy’s growing at a full-employment rate (these days judged by the econocrats to be an unemployment rate of about 5 per cent), it’s just not possible to ‘create’ any additional jobs, merely to move existing jobs from one location to another. Disinterested economists never cease to be amazed by some of the claims made about how many jobs will be created - or, with other lobbying propositions, lost.
