Andrew Leigh

Ten Lessons for Economic Policymakers

Economic Society of Australia Annual Dinner 2023
Commonwealth Club, Canberra

Wednesday, 1 November 2023

Introduction: The Power of Ideas

[Acknowledgements omitted]

John Maynard Keynes once wrote ‘The ideas of economists and political philosophers, both when they are right and when they are wrong are more powerful than is commonly understood. Indeed, the world is ruled by little else.’

In academia and parliament, I've certainly found that to be true. Economics is surprisingly powerful as a tool for public policy. Those of you who are established in your careers will know well the power that economics has had in terms of shaping Australia's trajectory.

Last month, we lost Max Corden, one of Australia's great economists, and somebody who, after fleeing the Nazis in 1939, became one of the great Australian pioneers of openness. Max's work on tariff reform was used by the Tariff Board, the predecessor to what is now the Productivity Commission, to make the case for Gough Whitlam’s 1973 tariff cut, in which all tariffs were cut overnight by 25%.

Max's story was one of coming to Australia, being welcomed here and becoming a great advocate for openness. He knew my grandfather, Keith Leigh, who died two years before I was born, and would tell me about how the two of them spoke of world events at Melbourne University in the 1950s and 1960s. That intellectual curiosity and global outlook reflects the very best of Australian academia and the economics profession.

You may have heard Thomas Carlyle’s put-down of economists as being ‘the dismal science’. Perhaps you know that the reason that Carlyle described our discipline as the dismal science was that we had what was in his mind the ‘dismal’ view that all human beings – whatever their skin colour – should be regarded as equal.

In that light, I proudly wear the badge of the ‘dismal science’. It is a reminder that economics has its origins in the notion of human equality; the principle that one person's wellbeing is as valuable to society as another's.

Max Corden was also a remarkably generous soul in terms of the time he spent with others. He always seemed to have time to ask junior researchers about their work. When I visited Melbourne University in 2006, I loved the chance to engage with Max, to chat with somebody who had worked on the world stage on issues of trade liberalisation.

My speech tonight proposes ten lessons for economics policymakers. When I refer to economic policymakers, I’m drawing a broad net. I'm including people who have made a contribution in consulting, those who have worked in the public service, those who are working in journalism, and those who contribute to the public debate. I'm thinking of the policy conversation writ large, not simply some narrow slice of it.

Ten Lessons

Lesson one: focus on wellbeing, not just dollars. As Bobby Kennedy famously noted ‘the gross national product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials.’

The notion that we teach first year students is that economics is about maximizing ‘utility’, also known as happiness or wellbeing. It's why the Australian Government has brought down a Measuring What Matters statement, grounding the objectives of government, not just in national income, important as that is, but also building out a wider dashboard of indicators.

One way of thinking about that is to imagine that you could only have one of the two developments that Australia has witnessed over the last half century. Imagine you could only have either a decade's more healthy life or the doubling of per capita incomes. Which one would you choose? I’d choose the extra decade of life and I know plenty of others who would do the same. If you make that decision then you're not just about maximizing GDP per capita. You're thinking also about the number of Christmases you have to spend with your loved ones.

This matters too when we consider policies that might boost employment rates. We know that the wellbeing of someone who loses their job goes down by more than you'd expect if a job was just an income delivery mechanism. The drop has been estimated by economist Nick Carroll to be somewhere between $40,000 to $80,000 in addition to the loss of income. A wellbeing approach better captures the social value of full employment. Wellbeing also matters in the domain of mental health. As the Australian Government works to build a more connected community, wellbeing is one of the things that drives us.

Lesson two: think comparative advantage. My guess is that most of you don't fix your own car when it breaks down, don't cut your own hair, and don't make your own clothes. What goes for the labour market can apply to international trade. Australia constitutes only about 0.3 percent of the global population. So it makes sense that we can benefit the most when we focus on what we do the best.

Just as comparative advantage suggests that you might be better off spending another hour in the office and paying someone to cut your hair, so too it suggests that Australia may be better off to focus on what we do well, whether that's extracting minerals or advanced production of pharmaceuticals, and then importing things that other countries produce more cheaply, such as smartphones.

Comparative advantage was cited by MIT’s Paul Samuelson as the best example in the social sciences of a proposition that is both true and non-trivial. You'll meet plenty of very smart lawyers who seem not to understand at all why Australia shouldn't produce everything domestically. That demonstrates the power of economic thinking. Comparative advantage embodies the notion that if we stick to our own knitting, that we can enjoy higher levels of prosperity, and higher levels of fulfillment.

To try and do everything is to turn ourselves into the human equivalent of a Swiss Army Knife with a funny little blade that doesn't really cut anything very well, a screwdriver that always seems to flick back, and a pair of scissors that will barely cut through anything at all. If we recognise the value of comparative advantage in our own lives, we can also recognise the benefits for Australia of comparative advantage.

Lesson three: ignore sunk costs. It is tempting to continue to throw good money after bad, to feel once you've engaged hard on a project you should see it through to completion. ‘Winners never quit and quitters never win’ goes the saying. Economics teaches us that very often quitting is exactly the right strategy.

A lovely experiment that Steven Levitt and Stephen Dubner conducted online invited people facing a major decision to have Levitt and Dubner toss a coin to decide whether or not they should quit. The only requirement was that you told them afterwards whether you'd followed the coin toss, and then reported how happy you were.

The research found that the biggest decision was whether or not to quit a job, and that when the coin came up leave, people were a little more likely to leave and as a result of that were significantly happier. It turns out that if you're on the verge of quitting, then it makes sense to quit. Why might it be that we stay on for too long? Most likely, because non-economists forget the notion of the sunk cost. People forget the idea that often, we may have invested in something which turns out to be a mistake.

Lazy firms maintain projects long after they've exhausted their value. More rigorous firms recognize that shutting down a team can be exactly the right thing to do. Astro Teller, who runs Google’s moonshot innovation unit, has a strategy of celebrating projects that are shut down. Here’s one of his stories: ‘Not long ago a team of 30 engineers killed a project they’d been working on for 2 years… The next week, I stood the entire team up on stage at one of X’s all hands meetings, and I announced they were getting a bonus for killing their project. I said, “Thank you! By ending their project, this team has done more to speed up innovation at X this month than any other team in this room.” And they got a huge round of applause from their managers and peers. They then went on vacation, came back, and found new roles in other teams around X, enabling us to go even faster at other ideas with brighter futures.’

Lesson four: be bold. Just as the most powerful ideas in economics are the foundational ones, the biggest value from our discipline can come from being the first economist in the room. This sometimes be a lonely role, but is often a critical one. To bring economics thinking into a non-economic discussion can completely transform the conversation.

It is said though, that this doesn't extend all the way through. The standard rule of thumb my economist friends have is that when more than 25 percent of the guests at a dinner party are economists, the conversation is immediately ruined, because the economists dominate.

So your boldness is best exercised in a context of a bringing economic thinking into a non-economic discussion. That can mean applying economics to policy problems that people think of as non-economic. To be part of a non-economic unit can be to bring a completely different perspective into the conversation, which may allow organizations to think in terms of cost-benefit analysis rather than a less robust framework.

Lesson five: use rigorous evidence. In building the case for change, it's important to recognize that the evidence in some cases may be very strong and in other cases may be very fragile.

Since dinner is now being served, let me use my very favourite example of this: nutritional epidemiology. For many years, newspapers and television and radio have traded off stories along the lines of ‘people who consume food X have health outcome Y’. That might well be nutrients and  cancer risk. It might be a particular vegetable associated with longevity. It could well be a particular beverage associated with mental wellbeing.

The problem, as health writers Peter Attia and Bill Gifford point out, is that our food choices are extraordinarily complex. Most likely, a study that looks at people who consume a certain kind of food is going to tell it tell you a lot about those kinds of people, and very little about that type of food.

Let me give you a particular example. Nutritional epidemiology found for many years that people who consumed moderate amounts of alcohol had better health outcomes than those who consumed no alcohol, or a lot of alcohol. This turned into a recommendation that people should have one glass of red wine a day, that would be protective for your heart.

But then a more rigorous form of evidence came along. This used the fact that on a random basis, a small portion of the population have a genetic intolerance to alcohol. Using that approach, known as ‘Mendelian randomization’, researchers could compare non-drinkers with drinkers. In those studies, non-drinkers appear to have better health outcomes than moderate drinkers.

Why might the nutritional epidemiology results have been wrong? Well, one reason was that it turned out that some of the non-drinkers were former alcoholics.  Moderate alcohol consumers also appeared to have better health behaviours in other regards. They tended to exercise more, they tended to have better mental wellbeing. It looks now as though the low-quality evidence was driven by the kinds of people who had the discipline to consume small amounts of alcohol every day, rather than by the causal effect of alcohol on health.

One of the things we're doing with the Australian Centre for Evaluation is to try and get more randomized trials into public policy, to move up the evidence hierarchy. As evidence improves, we're able to make better policy decisions. As economists, great use of evidence is one of the signatures of what we do.

Lesson six: consider expected value. Expected value is unfamiliar to most non-economists. Non-economists very free frequently think about things that will happen or won't happen. But economists are comfortable in the world of probabilities, and the world of risks in thinking about things that are unlikely to happen but aren't zero probabilities.

When we insure our homes, we're not buying a policy based on the assumption that our house is going to burn down in the next year. We're buying a policy that we believe is good value because the cost of the home burning down would be catastrophic if it were to happen.

So too we should use expected value calculations when thinking about risks to humanity. If you ask artificial intelligence researchers about catastrophic risk, the median researcher believes the probability that the long-run effect of advanced AI on humanity will be ‘extremely bad (e.g., human extinction)’ is 5 percent.

Those who don't think in terms of expected value might say that 5 percent is a small number. Those who do think about the expected value, think of the value of the world multiplied by 5 percent and come up with a very big number. If we can reduce the catastrophic risk of artificial intelligence risk from 5 percent to 1 percent, then that would have a huge expected value payoff for the for the world.

As I'm giving this speech, there is a meeting taking place in Bletchley Park in the United Kingdom, in which Prime Minister Rishi Sunak has drawn together a range of international policymakers (including Australia’s Richard Marles and by Ed Husic) to discuss how to reduce the catastrophic risk of artificial intelligence. Expected value tells us that the payoff for doing so will be very large.

Lesson seven: think in magnitudes, not just in signs. One of the things that struck me when I moved from academia to politics was the tendency of people to ask ‘will it happen or won't it happen?’, or ‘will it go up or will it go down?’.

But of course, the size of those effects are incredibly important. Thinking about the impact of having that daily glass of red wine on your health, I can reassure you that it is a very small impact compared to the risk of taking up smoking. If we're considering the impact of tax changes on behaviour, then fundamentally the most interesting question is not will a tax have a deadweight cost, because they normally do, it’s how big is that deadweight cost. In the case of insurance taxes and income taxes, both have a deadweight cost. But the deadweight cost of insurance taxes, according to most economists, is much higher than the deadweight cost of income taxes.

To focus on magnitudes is again to effectively use evidence. And by moving the conversation from up or down to big or small, we're making an important economic contribution to improving the conversation.

Lesson eight: channel a libertarian. Economics has origins in free market thinking. In 1776, Adam Smith wrote ‘It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest.’ Early French economists continued the focus on market efficiency, coining the term ‘laissez faire’.

In the same spirit, it can often be useful in any public policy conversation to ask: ‘what is the market failure?’.

You don’t have to be a libertarian to ask this question. Even economists who favour an expansive role for government should think about where the market failure exists. A valuable example is John Quiggin’s 2019 book Economics in Two Lessons, a response to Henry Hazlitt’s 1946 book Economics in One Lesson.

Hazlitt argued that economics can be boiled down to one simple lesson: market prices represent the true cost of everything. Quiggin offers a second lesson: market prices often fail to reflect the full cost of our choices to society as a whole. Quiggin then explores two-lesson economics by looking at market failures in a range of policy issues, including climate change, unemployment, monopolies and inequality. Quiggin is no devotee of Ayn Rand, but he nonetheless asks the libertarian question ‘what is the market failure?’, and in the process produces a valuable book about economic policymaking.

Lesson nine: remember equity. If you accepted lesson number one – consider wellbeing – then it likely follows that you care about equity. One of the best cases for redistribution is that most economists believe there is a declining marginal utility of income. To form this view, you need to simply believe that another dollar buys less happiness for a billionaire than for somebody who's sleeping rough. If you believe that, then it follows that redistribution can raise overall happiness. In a nutshell, you have the case for a progressive income tax and a targeted social safety net (for which Australia's is the best example in the world). Reducing economic inequality has the potential to boost overall societal wellbeing.

By considering equity too, we can ensure that we're not making the mistakes that flow from only looking at averages. If Elon Musk walked into this room, the average wealth would be over a billion dollars. But each of us wouldn’t become billionaires. To simply look at averages is to miss the distributional consequences. To miss the distributional consequences is to miss the welfare consequences. Even the most free market economist should focus on distribution, because to focus on distribution is to come back to the roots of economics in boosting wellbeing.

Lesson ten: choose openness. Economics has a history of favouring free trade, whether it's John Stuart Mill's campaign against the British Corn Laws, the open letter against the Smoot-Hawley Tariff Bill in 1930, or the world that Keynes and others did establishing the Bretton Woods Institutions, and slowly unwinding the protectionism that had emerged in the interwar period.

It isn’t just about trade. Economists have also been crucial in making the case for freer flows of money. Economic thinking encourages countries to be open to foreign investments. If you believe in investment, you should believe in FDI. Foreign investments means that we have more jobs and better paid jobs. Investing overseas also allows Australian superannuation funds to diversify their holdings.

Economists have also been keen advocates of migration. Our nation benefited from insights of Max Corden, who left Germany at the outbreak of World War Two. Australia has benefited from the millions of migrants who've come here and helped to build up the country in the post-war era.

When a migrant moves, they boost their own wellbeing, and generally improve the wellbeing of those around them. As economists know, migrants don’t just represent mouths to feed. They represent muscles to build and minds to inspire. To choose openness is to make the case against isolationism and autarky. In the face of rising threats from autocracy and populism, it is to believe that migration, trade and foreign investment can boost wellbeing.

Conclusion

So there are my ten lessons for economic policymakers, from the perspective of half a dozen years as an academic economist and a bit over a dozen years in parliament:

1. Focus on wellbeing, not just dollars.
2. Think comparative advantage.
3. Ignore sunk costs.
4. Be bold.
5. Use rigorous evidence.
6. Consider expected value.
7. Think in magnitudes, not just in signs.
8. Channel a libertarian.
9. Remember equity.
10. Choose openness.

In closing, I want to thank each of you for being part of the Economic Society of Australia. The Economic Society does a great deal to strengthen the intellectual conversation about economics, but also to build the social ties. In that sense, it embodies a couple of my rules. To be an effective economic policymaker isn't just about being bold in ideas. It's also about improving your utility and your wellbeing through the stimulating conversations you'll have with one another.

Thank you.