At the Fourth Annual Conference of Government and Economics, held virtually on April 26, 2022, audience members viewed a lecture from Jean Tirole on the role of government in the 21st century economy, with a particular focus on the balance between government and independent agencies as well as between government and its citizens. What follows is a transcript of his lecture that has been lightly edited for clarity. Jean Tirole is a 2014 Nobel Laureate in Economics and Honorary Chairman of Foundation Jean-Jacques Laffont – Toulouse School of Economics.

Thank you so much, David, and thank you to SAGE for inviting me. It's a great honor to be part of this conference. I'm going to share with you a few thoughts about the role of government in the 21st century economy.

I'm going to argue that the role of government is really a matter of balance, a balance between various things. And I'm going to start with the division of labor between the political side, which is the executive and legislative, as something which is more independent from politics, so the independent agencies, and ultimately the private sector of course. It is a big question in the West about whether the state is back, and to some extent it is.

There is a renewed popularity of industrial policy, and also plans for re-shoring. Of course, after COVID, there was some disruption of supply chains. Some, I'm saying, because actually they held up pretty well given the circumstances. But of course, now with the invasion of Ukraine, we also have the disruption of supply chains.

And of course, this is not the most important part of the world in this regard, but they still, if you think, e.g. about neon, which is used for chips, 50% of it is being produced in Ukraine. Now, the re-shoring, is that a short-term reaction or a permanent trend? I just want to point out that there are alternatives to re-shoring.

One is geopolitical diversification. The other one is, of course, the storage of essential inputs if we only have a short-term crisis.

And the second point I want to make is that there are limits to re-shoring. The first is that you cannot have access to what the entire world has to offer to you, and therefore there is a loss of welfare, a loss of purchasing power. The second point is that the creation of international supply chains is very much like a fixed cost, and once it has been spent, it's sunk.

For example, for Western companies to invest in China, it was a big investment. But now this investment has been made, so that suggests that maybe there will be some re-shoring, but not a huge amount.

My vision of the modern state is that I see the state as a regulator, a referee, and it's going to correct market failures. On market failures, I refer you to my book, Economics for the Common Good. You can think of that as being, for example, externalities, e.g. carbon emissions, and try to make sure that the general interests coincide with the individual interests and reduce emissions.

It can be market power, that's why we have competition policy. It can be the consumer's lack of information, that's why we have food safety regulation, or banking regulation. That could be inequality, of course, because there is no reason why the market is going to deliver the proper distribution of income and wealth and so on and so forth.

So it's basically a market fixer and a regulator or referee. Of course, the state has to be an enabler by providing various infrastructure like the legal framework. So good legal framework is, of course, very important for economic growth.

Where I would be more reluctant is to have the state as a producer, because in that matter, it's less efficient than the market and the private sector. Let's see why. State-owned enterprises, SOEs, are often inefficient for several reasons.

The first is that they are often given a fuzzy mission, so what they are trying to optimize is not completely clear. They have conflicting objectives. For example, they are being asked to be profitable, but at the same time, they may be asked to support the local industry or do various other things.

The second danger, of course, is that the government may actually select buddies for the top corporate jobs as a reward, but, you know, these may not be the most appropriate appointments.

The third issue is the so-called soft budget constraint. This means that if a state-owned enterprise loses money and it's about to go bankrupt, then in general, the government is going to bail it out.

And of course, the anticipation of bailouts is going to create a moral hazard, so the managers are going to be less concerned about losing money, and they will put less effort into avoiding lawsuits.

And finally, the state-owned enterprises may, in some cases, lobby the government so as to have no competitors at all, because, of course, a monopoly is a quiet life.

Another remark I want to make is that there is a loss of the power of the state in France and Europe. This has occurred over the last 40 years. Before that, there was actually a gain of power by the state, which went together with a very welcome rise in the welfare state. So we built a welfare state, and therefore the state became more important.

There were also some other reasons, like there were some nationalizations after World War Two, especially in France. However, from the 1980s on, there was what I would call a triple whammy.

First, there was some privatization of those state-owned enterprises. Then there was globalization with lower costs of doing trade, with trade agreements, starting in the eighties, nineties, and so on. And that globalization, of course, created more competition. The market became more important at the expense, a little bit, of the state.

In Europe, you had some other reasons for why the state lost importance. In France, for example, you had the devolution, in 1982, of power to regions. Everywhere in Europe you also had the European construction, which meant that some of the power went to the operation. And finally, we had a different state, which was the rise of the independent agencies. I'm going to ask, why? Why did this happen?

So to do that, we need to open the state's black box. The state is a big thing, and here at the top is some kind of constitution. And the constitution is really about the broad principle, not very detailed, just the values of society and what we want to achieve through the intervention of the state. It's the spirit of policymaking.

Then the details, which are extremely important of course, the letter, is determined by various actors. 'Co-opetitors,' with checks and balances, are going to fill out the details. I'm going to explain why I call them 'co-opetitors,' because basically they have to cooperate at the same time, but they do not have necessarily the same objectives if they want to implement the checks and balances.

Now, having checks and balances, in general, has a stabilizing effect on public decisions, so there are not huge swings when the government changes. And it creates some competition in politics, because absolute power may corrupt.

Monopoly, wherever it comes from—it can come from elite capture (there are lots of accusations of that in the West), or it can come from the party, or cliques, or whatever—generally a monopoly generates a feeling of superiority and impunity, and also groupthink.

Groupthink is when everybody thinks the same and nobody dares to present any contrary argument. And that's bad, of course, because after a while of groupthink, you end up with very bad decisions.

There's some balance between the legislative and executive branches, and of course, there is also some balance between the civil service and the politicians. And there we have a long-standing debate about the difference between the civil service and the civil servants.

The civil service refers to a long-term institution that serves the public interest. A civil servant is supposed to serve the interest of the government of the day. Of course, the government of the day in every country is going to say, 'I'm serving the public interest.' But of course, there is no reason to believe that. It's not quite the case, and we have to be careful with the difference between civil service and civil servants.

But in any case, what we need is some competition. So we need some cooperation, otherwise nothing gets done. But at the same time, the civil service and the legislative branch should not be 'yes men.'

Now, that brings me to a different conception of the state, which is independent agencies. That's not a new conception. If you think about justice, for example, the basic principle of justice is that the government cannot be judge and jury at the same time.

You need someone independent, and probably one of the examples was the British Crown. The judges in England became independent, in the early 18th century, from the British Crown. But of course, this idea that you need independent players in government extends. For example, that has been very important to develop anti-trust without intervention from ministers in favor of some firms.

That has been very important for regulation, for industrial policy agencies, for central banks. In most countries in the world, now, central banks are independent, and that's how we were able to tame inflation, for example, and also to do better banking regulation. And we want to avoid the prevalence of lobbying and electioneering rulers' interests into those decisions.

Now, independence, in general, facilitated protection. It protects the agency from the temptation to pander, to please someone. So basically, to be more faithful to the mission, whether it's central banking or competition policy, or what not.

That basically makes it easier to fulfill the mission and to have more expertise-based public policy with a lot of transparency. Because the managers of the agency are not politicians, they are not elected or anything. And therefore, they can explain very clearly why they took this decision and have a consultation, very open, where people who disagree express their views, and so on. And that's good.

Now, I have to warn you, an independent agency—and I like independent agencies—it's only as good as the quality of its staff. In an independent agency, if you put the wrong staff, it is not a good thing. So you have to find people who have expertise, who are independent from the government and from lobbies, and are honest. The selection process is very important.

And the independent authorities also are never completely independent in that they make a decision and it cannot be challenged. Over the years, if it doesn't fulfill the mission, then you could have, for example, the Parliament remove the head of the agency.

Now, let me come to the second part of my talk. What creates trust in government? Now, I'm going to look at a different kind of balance between the state and its clients, its users—the citizens and companies. And I'm going to make four suggestions for building trust.

The first is kind of obvious, it is to achieve bureaucratic efficiency. The state is there, at the service of the people. And of course, when it's business, of corporations, but it's basically at the service of the people.

So it has to achieve a low cost of doing business and a low cost of being a citizen. That implies that you should have simplicity of use. Therefore, it should be very easy to implement the various regulations that are designed by the government. For employees, households, and companies, a maze of support mechanisms or regulations can have very negative consequences.

One of them is obvious, which is that we may have to spend a lot of time understanding and trying to take advantage of the rules. So in France, for example, but in many countries, you try to find a grant for everything you need to finance—universities, businesses, and so on.

Conversely, there can be discouragement, because if it's too hard to find out how to do it, then there may be a low take-up rate of the benefits. And it's not a fair mechanism anyway, because it rewards people who do have information, so we don't want that.

So, we have to achieve bureaucratic efficiency. We have to also avoid a bloated state. A bloated state involves overstaffing—too many people doing the job.  It also involves the swelling of administration relative to operations. So, for example, in a school hospital, you may have too many people doing administration, and not enough people teaching, or surgeons, or the like. Another characteristic of a bloated state is the duplication of agencies or local authorities having similar prerogatives. So, we need to avoid all those things.

I have a few suggestions here, and they are not new—they are kind of obvious. The first is to evaluate, evaluate, evaluate, the efficiency of the public sector. And of course, to do that, you need auditing agencies that are independent of political power. This is very important because a state cannot be judge and jury at the same.

In the evaluation, you can use benchmarking. So, you systematically compare the government of one country with other countries. For example, how many teachers do they have? How much do they pay their teachers? And can you do better and get better education for your children? Between regions, between agencies, between universities, and the like.

So, for example, if you are a Western country, and you want to do a reform of the state, you can compare with the success stories of Sweden, Canada, or Australia. Sometimes to create benchmarking, you need to experiment. So if you have a uniform policy, that prevents comparing and seeing what works and what doesn't work. China in the 80s and 90s actually did that right, because experimentation allowed China to see what works and what doesn't work.

And you have to share. This information is not going to go into a drawer. You need to share the evaluations with officials and citizens as well.

The second recipe, in a sense, is that you have to ensure fair treatment. There is a frenzy of regulatory loopholes in many countries. So, for example, some people pay a carbon tax, some people don't pay a carbon tax. You have different pension schemes, and you have lots of tax loopholes. Some people pay their taxes, others don't. You can have protection of competition in some industries and so on.

This is both inefficient and unfair. It's inefficient because you don't meet the objectives. You don't reach the objectives that you have designed, but also in terms of political economy, that means that everybody is going to lobby the government to have access to this exemption or obtain an override. It's unfair, and it's going to create a feeling of inequity for those who do not have access to the loopholes.

So, in principle, you have to avoid hubris. You have to be humble when you are in government. And that's normal, because government officials don't have all the information that is needed, and regulation is largely an issue with information that you have.

The consequence is that you have to realize that you don't always have information, and you should not design public interventions that are dependent on data that the public authorities do not possess. Otherwise, you are going to create cronyism and arbitrariness.

Actually, insufficient information is why we are using so many, what I would call, untargeted policies. For example, R&D subsidies that go to every firm, experience rating for layoffs, or a uniform carbon price, which all economies almost recommend.

But sometimes we want, nonetheless, to use finer information and design some kind of industrial policy that favors and supports specific industries, specific technologies or firms, and picks a 'winner,' but then it's very crucial to adopt the proper governance.

In my book, Economics for the Common Good, in chapter 39, I list the best practices for industrial policy. You must use independent and qualified experts to select the project for public funding when you do industrial policies. You have to bring in the experts, the scientists for example. And they have to be independent—they are not there to please the government, they are there to make things happen.

You also have to pay attention to the supply side. You must see, 'if I spend money there, I'll say the people wanted to make it happen.' So you have to plant where the soil is fertile.

Next, you should not prejudge your solution. If you go back to COVID and the vaccines, when we thought about possible vaccines, we actually didn't know which technology—old-style virus, or viral vector, or messenger RNA—would work. Nobody knew. And it turns out that messenger RNA actually worked very well, and that was a little bit of a surprise. So we should not prejudge a solution. We must have a goal, not selected technology.

We have to evaluate the intervention, publish the results, withdraw support if the project doesn't work or is no longer needed, and we have to involve the private sector in risk taking. Surprisingly, the role model there is DARPA-E. Now we interact heavily through computers, and a lot of that computer technology actually comes from fundamental research financed by the DARPA or DARPA-E. 'E' stands for energy now, but at the time, that was mostly about computer science. Other role models include grant funding agencies such as those for science in universities and outside universities. That's very important if you want to actually achieve good industrial policy.

My last point, which is very important, is that you have to address what I call 'time bombs.' What is a 'time bomb?' A 'time bomb' is a policy area where if you don't do anything for a year or two, it doesn't matter.

So, think about climate change. If you don't act on climate change for a year or two years, it doesn't make a big difference. Almost no difference in the short-term, and the entire difference will be in the long-term.

But of course, this is bad, in a sense, because it means that you have political procrastination with years or decades of inaction. Actually, in the case of climate change, we have had, all over the world (except in a couple of countries, like Sweden) basically inaction. Three decades of inaction because of this time issue.

And climate change is not the only example of that. If you think about education, it's the same thing. If you don't do anything for a year or two, it doesn't really matter, but of course, in the long term, it's a disaster. Higher education, R&D, public debt, health prevention, domestic and international harmony, avoiding repression and belligerent attitude—it's a short-term policy, it's not a long-term solution.

Now, there's this debate about whether democratic or autocratic regimes are more prone to create time bombs. The answer is: I don't know what the answer is. It's true that in democratic countries, politicians may privilege reelection, and basically not act. But at the same time, the autocratic regimes may be blinded by the absence of democratic opposition or NGOs. And actually, on the issue of climate change, you have a lot of procrastination in both camps.

So let me stop here and conclude. The governance of the state is very complex because it has a very fuzzy mission, and nonetheless, it's extremely important. The government may turn out to be extractive, or it may actually be a complement to economic activity. What I've tried to do is to share with you a few thoughts on good functioning, and I hope it has been useful.

I wrote a report with Olivier Blanchard last year, for the French presidency, but also for Europe more generally, on some major economic challenges. And of course, I wrote a book a few years ago, Economics for the Common Good. Thank you very much for inviting me, and I look forward to your questions.