Monday, September 17, 2001

WHAT IS TREASURY’S ROLE?

Talk to Treasury seminar, Canberra
September 17, 2001


I suspect that, when Martin and Ken invited me to talk to you today, their intention was to put before you a speaker you’d find a bit thought-provoking and challenging – someone who’d give you a bit of an intellectual workout. (Maybe someone you could easily dispose of before lunch.) So, to that end, I’m going to take a round-about approach to my topic of what Treasury’s role should be. I’ll be sneaking up on the topic, dividing my talk into three parts. The first part will be a bit theoretical, the second a bit philosophical and the last very down to earth. When I finally get round to talking about Treasury’s role, I’ll throw a few friendly punches in Treasury’s direction.

But let’s start with a little challenge on the theoretical side. If you’ve read my stuff over recent years you may have noticed that I’m much influenced by the insights of ‘behavioural economics’. This is a relatively new school of economic thought which is led by Richard Thaler – formerly of Cornell, but now ‘promoted’ to Chicago – and draws heavily on the work of two cognitive psychologists, Daniel Kahneman of Princeton and the late Amos Tversky of Stanford. If you’re not familiar with behavioural economics, you may have had more exposure to its sister discipline, behavioural finance, which is devoted to challenging and improving upon the efficient markets hypothesis. One of the leaders of behavioural finance is Robert Shiller of Yale, author of Irrational Exuberance.

Behavioural economics challenges two basic premises of the neoclassical model, rational choice and, under conditions of uncertainty, expected utility theory. These theories are both normative and positive – they specify not only how economic agents should act, but also how they do act. Behavioural economics subjects these theories to empirical testing, mainly by means of laboratory experiments, and finds that they fail as positive theories: they are poor predictors of how people actually behave. Let me summarise some of the key findings:

1. Money is not fungible. One dollar is not the same as another. In the minds of most people, a dollar’s value varies with circumstances. For instance, people will go out of their way to achieve a saving of $5 on the purchase of a tranny worth $25, but won’t go to the same trouble to save $5 on the purchase of a TV worth $500. People fail to treat all dollars equally because they engage in a system of metal accounting, where certain sums are allocated to certain purposes and can’t be drawn on for other purposes. They engage in this seemingly irrational behaviour because it helps them keep their total spending within their budget constraint. One key finding under the heading of fungibility is that a dollar of loss gets a higher weight than a dollar of gain.

2. People aren’t risk averse so much as loss averse. It’s true that people are risk averse when it comes to gains, but their dislike of losses is so great that many are willing to accept gambles in the hope of avoiding them.

3. People are more concerned about the change in their wealth than about its ultimate level. I offer you the choice between starting the day with $150 and suffering a loss of $50, or starting and ending the day with $100. Theory predicts you’ll be indifferent between the two but, in fact, most people would much prefer to avoid the loss by starting and ending with $100.

4. People simply can’t see why they should ignore sunk costs. This is something politicians understand much more readily than their econocrat advisers. It occurs, I suspect, because of people’s misconceived desire to avoid losses.

5. People often ignore opportunity costs. They commonly have a wide gap between willingness to pay and willingness to sell. I may not be willing to pay much for a particular item, but once I own it, you’d have to offer me a lot more before I was willing to sell it to you. This is a bias towards preserving the status quo.

6. People often fail to order their preferences consistently and have difficulty making comparisons between three or more things. As part of this, people’s choices are often influenced by the way those choices are presented to them – the way they’re ‘framed’ – which is why, for instance, we know not necessarily to believe the results of opinion polls conducted by vested interests.

What all this amounts to is an exploration of the specifics of Herbert Simon’s notion of ‘bounded rationality’. The point is not that people are wilfully irrational or capricious, but that the human mind simply has limited capacity to store and process information. We are physiologically incapable of being as coldly rational in our decision-making as the neoclassical model assumes we are. We rely on simple rules of thumb (or ‘heuristics’) which generally serve us well – they’re mentally cost-effective – but nonetheless lead us into consistent errors.

The fact that the leading proponent of this stuff has now won a chair at Chicago suggests it’s starting to gain some respectability in academic circles. My motive in drawing it to your attention is partly to give you something to think about and certainly to challenge any complacency you may have about the adequacy of the neoclassical model. The great stumbling block to the wider acceptance of the insights of behavioural economics is the difficulty in incorporating it in a mathematical model. If a truth can’t be turned into an equation then a mighty lot of academic economists would prefer not to know about it. But that’s not a good enough excuse for someone engaged in policy advising to elected politicians. Their advice needs to be tempered by the knowledge that the model’s assumption of rational choice is unrealistic and unreliable. Advisers need to understand and, where possible, anticipate the misconceptions that bind the minds of ministers and their constituents. They should appreciate the great extent to which rational responses need to be explained and justified. And it’s not enough to convince the minister; you then have to provide him with the explanations he can pass on to the public. The insights of behavioural economics are also useful in helping advisers realise when they’re pushing their luck too far – when the propositions they’re advancing are just too counterintuitive for any politician to sell.

By way of illustration, I’ve argued before that these insights do a lot to explain the political difficulties of selling tax reform. Once the politicians committed the fatal design error of making the reform of indirect taxation revenue positive rather than revenue neutral, the basic proposition they had to sell was along the lines: the net effect of the GST is to raise your cost of living by $10 a week, but don’t worry, we’re going to cut your income tax by $20 a week, so you’ll be $10 a week better off. To any economist, that sounds a pretty attractive offer. But to many voters it sounded most unattractive. Why? Because they gave the $10 of loss a higher weight than the $20 of gain.

In the outworking of the tax package, many voters believe their personal cost of living has risen by far more than the 3 per cent or so that the CPI seems to be suggesting. Behavioural economics’ exploration of the various heuristics that affect people’s thinking – such as that they find it easier to call to mind unusual or outstanding events – does much to explain why the CPI people carry around in their heads gives such radically different answers to the one the Statistician so carefully calculates.

OK, now let’s move on to the second part of my remarks, the vaguely philosophical bit. In his speech to the Fin Review’s chief finance officers’ summit in June, Martin repeated an argument often advanced by Treasury and others that the significant and sustained increase in productivity growth during the 90s can be attributed to the beneficial effects of 15 years of microeconomic reform. Though it can’t be proved, I have no difficulty accepting this argument. But it prompts what I regard as a key question: if we’re so rich, why aren’t we happy? If income per capita grew faster in the 90s than at any time since the 60s – and it did – why is there so much dissent in the ranks? Why so much support for outfits like One Nation; why so much abuse of economic rationalism, such carry-on about globalisation and such vehement criticism of national competition policy? Why do we all know that, no matter who wins the election, little further reform is likely to be undertaken?

Treasury’s standard answer – of which there were hints in Martin’s speech – comes in two parts. First, the opposition is coming from vested interests defending their privileges. Second, to the extent that this doesn’t explain the discontents of ordinary mortals not being whipped up by lobby groups, the punters are simply too dumb to understand what’s good for them. It’s usually put a bit more politely than that: the policies haven’t been ‘sold’ properly. I have to say the line that the public is woefully illiterate in economic matters – that it just can’t trace through causes and effects – doesn’t come well from economists whose advice is based on a model that assumes agents’ decision-making is relentlessly rational at all times. But, apart from that, the standard defence isn’t sufficiently convincing. It’s too self-serving, for one thing. And the line that the punters don’t know what’s good for them, so the pollies should just press on with good policies because, in due time, the results of those policies will convince people that father knew best – well, time’s run out on that argument. The benefits are in, they’re in line with the economists’ predictions, but the public’s still not happy with the outcomes.

The standard response from Treasury’s critics within the profession – from people such as Bob Gregory, let’s say – is that the reason higher income per capita hasn’t satisfied the electorate is distributional. Sure micro reform has increased national income as promised, but – not surprisingly – too much of that extra income has gone to the top 10 or 20 per cent of the distribution, and too little to the battlers. It’s no wonder the battlers are in revolt – and justifiably so. Now, you won’t be surprised to hear that I have a lot of sympathy with this standard criticism – and I’ll have more to say on it later.

But I want to give you my own, non-standard response – the philosophical bit. Bob Hawke once said something to the effect that the role of economists was to increase happiness. It struck the public – and even some economists – as a weird thing to say, but I think it’s right. I don’t see much distinction between ‘utility’, ‘satisfaction’, ‘welfare’ and ‘happiness’ (or, as the psychologists usually put it, ‘subjective well-being’). So, to me, increasing happiness is the stated objective of economics, and the question is: how well does it do in achieving that objective? As you may know, there’s a growing body of empirical literature about the relationship between income and happiness, and I believe it sheds a fair bit of light on that key question raised by the success of micro reform: if we’re so rich, why aren’t we happy?

A host of studies have shown that, though rising per-capita incomes have been associated with rising measures of subjective well-being in developing countries, in developed countries the steady rise in incomes over the past 30 or 40 years has not been associated with a measured improvement in subjective well-being. So we’ve been producing and consuming more goods and services but, contrary to the most basic assumption of economics, it doesn’t seem to have led to any increase in aggregate utility.

Why not? The literature offers various explanations. One is what Kahneman calls the ‘hedonic treadmill’. As our income rises, our expectations quickly adjust, leaving us feeling no better off. This is consistent with the findings of Richard Easterlin of USC in a recent study: ‘income growth does not . . . cause well-being to rise, either for higher or lower income persons, because it generates equivalent growth in material aspirations, and the negative effect of the latter on subjective well-being undercuts the positive effect of the former.’ People continually expect a rise in their income to make them happier, but it never does. They fail to anticipate the rise in their aspirations that comes with their higher income.

A second explanation of why higher national income fails to increase aggregate utility comes from the empirical observation that – contrary to the assumptions of the model – what people strive for is not an absolute increase in their income but a relative increase. What they’re seeking is socio-economic status – a higher place in the pecking order. So they’re keeping up with the Joneses – or rather, trying to get ahead of them. In consequence – and as Robert Frank of Cornell argues in his book Luxury Fever – the more affluent societies become, the more consumer spending is devoted to Fred Hirsch’s ‘positional goods’. If so, it’s obvious why we’re no happier: we’re engaged in a zero-sum game. Any increase in my relative income generates a negative externality for others.

If there’s anything in these arguments, they suggest that micro reform is pursuing an intermediate target – higher national income – which doesn’t lead to the ultimate objective: greater welfare. They also suggest that materialism is the great delusion – the great carrot on a stick – of our age. I suspect that when the punters carry on with arguments about how ‘people are more important than The Economy’, what they’re expressing, in their own inarticulate way, is their disillusionment with materialism.

In recent times I’ve become more conscious of the narrowness of economics. That it deals with just one, important but narrow dimension of our lives: the material. The limitations of the neoclassical model narrow it further. Like all professions, economists are highly specialised in their interests and they run the occupational risk of developing tunnel vision in the advice they give their clients. I’ve decided I don’t want to devote what remains of my career to trying to ram materialism down people’s throats. And I’ve become more conscious of my obligation as a journalistic commentator to warn my readers not to be unduly influenced by one-dimensional advice from economists – by what, to coin a phrase, is ‘partial-equilibrium analysis’. Voters have to stand up to economists in the same way an HSC student has to stand up to a French teacher who lays on the homework as though French was the only subject the student was studying. So the first thing I’d say about Treasury’s role is that, in proffering its advice, Treasury must guard against the unconscious assumption that the material objective should be maximised at the expense of all other objectives. I suspect Treasury also needs to be more conscious of the analytical limitations of its discipline.

So at last we reach the final part of my talk where I actually address the topic and say something about Treasury’s role. I’m not going to bother telling you what a great job I think you’re doing – there’d be little point in that. No, I’m going to offer some criticism. But please note this: I’m a person who earns his living by expressing his criticisms in public, but these are being offered in private. And though I’m paid to find fault, I have no reason to be anti-Treasury. In fact, I’ve always had in my makeup an instinctive sympathy for those people charged with the daunting task of holding the show together and moving it on in pursuit of the community’s betterment.

To that end, I’ve always accepted, for instance, that the Treasurer’s role is to talk up the economy – that he does so in our corporate interests, not just his own political interests. I’ve always accepted that it’s not Treasury’s role to be the first to forecast a recession, for fear of the self-fulfilling prophecy. I’ve been sympathetic to Treasury when its Budget-time forecasts have been subject to cheap shots from smart-alec business economists who enjoy the luxury of being able to change their forecasts as often as they change their underpants. (And I almost always remember to pin the forecasts on the Government rather than Treasury.)

Hitherto, I’ve always accepted the need for Treasury – and more, these days, Finance – to play the institutional role of Dr No. I think of it as the role of the cricket club treasurer. She’s facing a committee of 12, each of whom can think of fabulous ways of spending every cent in the club’s bank account – and more besides. Facing such odds, if she adopts a position of automatically opposing all spending proposals, there’s a fair chance that no more will get through than the club’s finances can stand. There’s never any shortage of people with good ideas to spend money. Every institution needs a least one person – the treasurer - who knows her role is to be a perpetual wet blanket in the interests of the institution’s longevity.

And, hitherto, I’ve always accepted the need for Treasury – and Tax, of course – to ‘protect the revenue’. If you don’t do it, who will? Well, it was my own concern to protect the revenue that helped to bring me into journalism almost 30 years ago. I was a young chartered accountant bored with auditing. I decided the only really interesting part of accounting was tax advice, but I also decided that I didn’t want to devote my life to helping the well-off avoid their responsibilities to the community. I trust you’ve seen this ‘prior’ evident in my work. I’ve often been happy to defend unpopular tax measures – I even went to the defence of the super surcharge, despite considerable reservations about its design features.

But – and now we get to the but – of late I’ve formed the view that Treasury needs to perform its roles of opposing new spending and defending the revenue with a lot more forethought, discernment and subtlety. With more guile, if you like. Let’s start with the spending side.

In Martin’s speech to the CFOs, he argued that ‘these real GDP per capita gains [from micro reform] are not just materialistic gains for the lucky few, but a real improvement in Australia’s capacity to sustain whatever economic and social improvements its people want’. I responded to that by email, and I’ll tell you my response. I said that, coming from Treasury, it was a bit rich. It was, in fact, no more than ex-post rationalisation. We both know, I said, that, first, when Treasury and other economic rationalists were urging governments to make these reforms in the first instance, they did little to urge governments to introduce (quote) ‘policies that help sections of the community adjust to change’ (unquote). In fact, Treasury hoped to be able to get away without them. Second, when other people wanted such adjustment assistance/compensation for the losers, Treasury oftentimes opposed it. The third thing we both know, I said, was that when the political shite eventually hits the fan, the pollies go behind Treasury’s back and do things that are really stupid. In which case, Treasury’s badly overplayed its hand.

Consider the case of the revolt in the bush. We’ve been through a period of at least a decade where, as a result of a range of micro reforms – from banking to railways to telephones – people have been busily dismantling a whole range of long-standing cross-subsidies in favour of RARAland. It seems to have occurred to no-one in any of the alleged co-ordinating departments that these disparate federal and state initiatives were starting to add up, and might soon provoke a counter-reaction. It occurred to no-one that there is deeply ingrained community support for the notion that the city should prop up the bush – support not just in the bush, but also in the city. When the inevitable backlash came, what did we get? We got petrified politicians running everywhere reversing policies and oiling squeaky wheels. What we didn’t have was any pre-emptive thinking by the PC or anyone else about second-best solutions. Thinking that, proceeding from the position that the public would always want some degree of subsidy for the bush, studied the costs and benefits of rival subsidies and came up with a program saying, if you insist on wasting your money, these are the least-worst subsidies to waste it on.

Martin always expected what I’d say today was that micro reform should have done more to help people adjust to change and more to redistribute the proceeds of reform from the winners to the losers. Well, I am saying it. I’m saying it’s not enough that the winners could compensate the losers; you actually have to do it. I’m also saying that if you want to know why economic rationalism and globalisation and competition policy are on the nose – and why the prospects for further reform are so dismal – you need look for no fancier answer than this. The punters have discerned that, whoever’s making a killing from micro reform, it ain’t them. And they’re right, it ain’t. The final ain’t is that they ain’t gonna take it anymore. Treasury and its fellow travellers sought to drive too hard a bargain on the opening up of our economy, and they’ve brought to whole game unstuck.

Next I want to say something about protecting the revenue. Alan Mitchell has a saying that the main task of the bureaucracy is to keep the crazy decisions to a minimum. On the tax front, I don’t think Treasury and Tax have been doing too well lately on that criterion. They seem to have been working on the theory that if you look after the pennies the pounds will look after themselves – and they seem to have come badly unstuck.

Consider the petrol excise. I assume that Treasury urged the Government to do the 1.5 cents a litre chisel on its promise to prevent petrol prices rising, and that it urged the Government to include the special GST price effect in the regular indexation of excises. These penny-pinching decisions were taken or confirmed after the price of petrol had skyrocketed and motorists’ extreme displeasure was apparent. Presumably, the logic was that, with OPEC and the Aussie dollar doing their worst, no-one would notice a little chiselling by the Government. But the Government was robbed of the ability to claim its hands were completely clean in the matter, and the motoring lobbies were incited to launch a major campaign. In the inevitable backdown, the Government not merely gave up the ill-gotten 1.5 cents a litre, it abandoned the indexation of petrol excise for good and all. That’s protecting the revenue? Well done.

Next, consider the alleged self-funded retirees. My mail says that a lot of the complaints Government backbenchers were getting about GST paperwork came not from small businesspeople complaining about the BAS, but from retirees complaining about the IAS. And why wouldn’t they? When the backdown came, you could have knocked me over with a Payment Summary – Individual Non Business when I heard Costello say that people likely to owe less than $250 a year no longer needed to complete a quarterly IAS. If having the elderly submit five returns a year to pay less than $250 isn’t administrative incompetence (ie they didn’t have to, but were allowed to think they did), it’s appalling penny-pinching. And the ultimate penalty to the revenue? The even more appalling measures to square away the self-righteous retirees in this year’s Budget – measures that are like a time-bomb ticking away as the population ages and Generation X buckles under the weight of carrying the babyboomers.

What is Treasury’s role? Well, one role is that, in its efforts to control government spending and to protect the revenue, it should avoid being too smart by half.

Read more >>