Robert Engle, a finance professor at New York University’s Stern business school, became a Nobel laureate in 2003. He won the prize for ground-breaking and highly complex work on the statistical concept of autoregressive conditional heteroskedasticity (thankfully abbreviated to Arch). His work made it easier for portfolio managers to determine asset prices and to model volatility. In particular, he has concentrated on measuring long-term risks for portfolios.

When it comes to long-term investing, Prof Engle’s most interesting proposal is a dramatic one. He wants to solve the problems of rising carbon emissions and pension deficits by levying a carbon tax whose proceeds would go towards a special, semi-autonomous pension fund.

Somewhat sceptical about the work of behavioural economists, he suggests that this year’s exceptional volatility is not about irrational behaviour by investors. Rather, he says, people rationally used economic models that turned out not to work.

JOHN AUTHERS: The most widely used measures for risk management, such as value at risk, tend to be very short run. How do you measure long-term risks and their potential downside?

ROBERT ENGLE: Longer-run risks are basically economic risks of one kind or another. It’s possible to link things like macroeconomic uncertainty to future financial market volatility, and, consequently, risks.

If you look at the long-term historical data, as you know, we’ve had long periods of really low volatility but other periods of high volatility, and the question is what causes these kinds of changes. Mostly, we would describe them as macroeconomic, sometimes they are about the financial sector.

JA: So how do you explain the sudden extreme volatility that hit markets earlier this year?

RE: My interpretation is that there are two effects that basically add together to give what we are seeing. One is more macroeconomic uncertainty. Every day as we read the paper we come up with individual estimates of whether we are more likely to be heading for inflation or recession. Is the housing sector going to pull us down, or the export sector pull us up? I think there’s more macroeconomic uncertainty than we’ve had for the last four years. That’s a basic natural determinant of increased volatility. These are long-term risks but we are doing our best every day to forecast what they are going to be.

The second [effect] is the whole credit issue. I think basically that every sub-prime lender must expect defaults. But somehow they are now forecast to be a lot higher than they expected.

JA: Why are people surprised by this?

RE: I would say there are three reasons. First, the historical data that’s used for risk management is too short to cover any other housing recession we’ve had. If you look at historical data it looks like it couldn’t have happened. Even in the early 1990s, there was very little subprime lending. That’s not going to give you good information. We’ve only seen a good housing market over the last eight years, and that’s how we are estimating defaults.

Second, there were a lot of lax lending procedures. And third, and most interesting, all those loans are securitised and it’s very difficult to know how risky they are. Rating agencies and investors have got them wrong. Analytic models that are used to price the tranches of securitised loans are really not adequate to reveal the risks. And so investors all over the world have taken a large amount of what they think of as sterilised subprime loans. But they aren’t. They would take on the senior tranches and a few extra basis points for what turned out to be fairly large risks.

JA: Does that mean we will see less securitisation in future?

RE: I think the securitised markets serve a very important function. They spread risks. I think they are here to stay. Possibly it will be less important. Certainly, I think everyone from rating agencies to investors are struggling with how they understand what they’ve seen and how they build better models for rating these things.

I don’t think that means that these kinds of securitisation will go away. They serve a very important economic function. But they lulled many people into taking a risk. That’s a little reminiscent of 1987, with portfolio insurance.

JA: Do you think behavioural finance, which looks at ways in which people are systematically irrational, could explain what has happened?

RE: I’m not convinced you need behavioural explanations to understand what’s happening. I think that they are particularly relevant for unsophisticated investors. But I think that within these investment banks, pension funds and so forth, people are making decisions on rational grounds which happen to be wrong. They were inaccurate models. That’s bound to happen.

JA: You’re well known for the principle of asymmetric volatility. Could you explain that?

RE: When things go down, there’s a prediction that volatility will be higher in the future. In fact, when you see a market descend, you naturally would forecast that the future volatility is going to go up a little more than you would expect given the size of the market move. People have always wondered why it’s true.

I think it’s a natural result of sensible investment strategies. As soon as you get some information that the financial markets are going to be volatile in the

future, you will reduce your exposure. And that means that the prices have to fall a bit to induce people to hold the same amount of assets. That means prices discount in advance. Volatility tends to go up in bear markets.

JA: So is there any way to profit from this insight?

RE: I don’t think there’s a way to benefit from it. You are doing your risk-return trade-off. It changes your allocation. And by adjusting your allocation, you are doing exactly what creates the asymmetric volatility.

JA: It’s plain that the fund management industry faces severe long-term problems. How do you suggest solving them?

RE: If you think that there is some probability that the climate will be pretty bad in 20 years, and higher taxes will be needed to do whatever needs to be done, then the economy as a whole is likely to be less productive. That’s a long-term risk like a business cycle risk.

If you think that the investment you are buying today is facing that risk, you might pay a little bit less for it. So this could have an effect on asset markets today.

I think it’s the same issue with pension reform. Do we think that in 20 or 30 years there aren’t going to be enough workers to support all the retired people? We are going to see higher payroll taxes, or else reduced pensions. This is a future risk that actually doesn’t look like it’s uncertain. It’s really going to happen.

My proposal is that you treat these two as part of a package in implementing the following policy. You pay a tax on carbon emissions, which is what most economists think is the simplest, if politically least attractive, solution to the problem. But instead of taking the revenue and doing something like spending on alternative fuels, you put it into a fund, the way Norway and other countries have done [through Norway’s sovereign wealth fund, which has more than $350bn in assets]. It’s an investment fund that accumulates value over time and it would be used to support pension commitments in the future. They have the same kind of time frame. Instead of raising taxes on labour, which they would rather not do. That shifts the time of payments for the pension fund to the generation while they are working. It’s more like a pay-as-you-go kind of system.

Second, it shifts the burden of the tax to being a benefit rather than a cost. You offset those two. My hope is that the idea that you can save both the climate and the pension fund by one tax might make it politically more acceptable.

JA: That’s an ambitious proposal. Who would control this fund, and how would it invest?

RE: I would think you would want it as a semi-autonomous fund like the Federal Reserve Board. They would report to Congress what they are doing. But the government can’t actually extract funds from it.

JA: These are both very long-term issues, and it would take a lot of political argument in the short term to make it happen. Would there be any short-term benefits?

RE: It does this in a way that doesn’t leave it up to individual decisions. If you can ameliorate those long-term risks, you will see improvements today. People will say: “It actually looks like the companies I invest in will have a better long-term outcome. Therefore, you would see the benefit of improved economic performance now. So it’s not necessarily just this big cost. It’s just a shift of taxes.

And would individual investors have any control over the fund’s investment choices?

Obviously individual choice does make some sense. But here I’m proposing a semi-autonomous government agency to make those choices. It would probably mostly be doing some sort of indexing. Otherwise it becomes more complicated.