[MUSIC PLAYING] GILLIAN TETT: We now have what promises to be an equally lively discussion about investment theory, financial valuations, markets, and the economy. And we're going to hear-- we're lucky enough to hear-- from two veritable luminaries, if I dare not say, lions of the financial economics world over the last few decades.

A couple of years ago, my colleague, John Autherrs, who is the investment editor of the Financial Times, wrote a piece pointing out that Robert Shiller, professor at Yale, Jeremy Siegel, professor at Wharton, essentially encapsulated two very different perspectives on what was happening to financial markets and investing. One is a bull. One was a bear.

One essentially believed in the CAPE theory. One did not. He wrote that piece, my colleague, two or three years ago. And back then, he pointed out that US stock markets were very high. And depending on which perspective you believe, Professor Shiller's or Professor Siegel's, you essentially had very different views on the valuations of the markets. Well, today, we have them both back again to talk at a time when the markets are even higher.

And what they have to say about valuations and investing is going to be absolutely fascinating about where you think equity markets might be heading next. The men barely need any introduction at all. They are, as I say, very famous, and it's an extraordinary treat to get them both on the stage. But let me start by introducing Jeremy Siegel. I think we get a nice drum roll and a picture.

[MUSIC PLAYING]

[APPLAUSE]

Glad to have you. Thank you. It really does feel like American Idol, because we have two competing candidates here with different views. But Jeremy Siegel, as I think is known to all of you, is the Russell E Palmer Professor of Finance at the Wharton School at the University of Pennsylvania. He is the author of numerous books, and numerous articles, and two books-- Stocks for the Long Run and Futures for Investors.

He is an academic director at the Security Industry institute. And he's the recipient of a Distinguished Leadership Award from the Security Industry Association and many, many other distinguished awards. He has a PhD in economics from MIT.

Secondly, we have of course, Robert Shiller, who is the Sterling Professor of Economics at Yale University. I think we have another drum roll.

[MUSIC PLAYING]

He also has numerous awards, including a Nobel Prize in Economics, and he's written numerous books, including Irrational Exuberance, and Finance, and the Good Society. So we're going to start with some comments from Professor Shiller about how he sees the world developing now. And then we'll turn to Professor Siegel for his views and his, if you like, differences. Thank you.

ROBERT SHILLER: Thank you, Gillian. So my disclaimer first-- I should disclose that I now have investment products through Barclays Bank that use the CAPE concept. That's all I'm going to say about that.

GILLIAN TETT: You're eating your own cooking.

ROBERT SHILLER: So this is also f 50th anniversary of my meeting Jeremy Siegel. And that photograph was taken almost 50 years ago at a party at MIT. So you recognize us? It's not the best picture, but it's the oldest picture I can find. So over the--

GILLIAN TETT: Do you remember that moment, Professor Siegel?

JEREMY SIEGEL: We had many moments like that, yeah.

ROBERT SHILLER: So over the years, I've learned many things. We've got into many arguments as well. And we've been arguing a little bit lately about the valuation of the stock markets. So I'm going to just launch right into that. Working with student of mine, John Campbell, in 1988-- that's 30 years ago-- we define the concept of CAPE, Cyclically Adjusted Price Earnings Ratio.

It's very similar to a price-earnings ratio except it's real price divided by a 10-year average of real earnings just to smooth out earnings in the denominator. It's not actually original with us. This sort of thing has been done for over 100 years by some analysts. But we said, this is really the right thing to do, because earnings are too volatile on an annual basis. When you're using the denominator to denote the value, fundamental value, the CAPE ratio varies all over the place.

But it seems to be stationary historically. When the CAPE is high, you might think it's portending future earnings growth. But Campbell and I found out it doesn't do that historically. It predicts future price declines. Let's look at the CAPE ratio.

The left chart on it is a price-earnings ratio-- 12 months trailing earnings-- back to 18-- I think it's 81 in the picture. The CAPE ratio is on the right diagram. And you can see it looks a lot different. You can see the noise from year to year in earnings, which just jostles the price-earnings ratio. If you want to get a measure of fundamental value of a company, you want a smooth earnings.

And we've done that. And we found-- Campbell and I found, that we could forecast the stock market. You can't forecast earnings using CAPE. But you can forecast price using CAPE. So now the question is-- there's other ratios. And I've been working-- these charts come from work that I've been doing with some of my Barclays Bank colleagues-- [INAUDIBLE].

And in these charts we see comparisons with other ratios. So these are so-called box and whispered charts. The red at the middle is the average. The extremes are represented. And today's value is represented. So CAPE is on the left. The price-earnings ratio is to the right of it. You can see that today's value on CAPE is higher than it is relative to history than it is with the ordinary price/earnings ratio, prompting alarm at the high-priced market.

But it's also true for the price-dividend ratio. Look how high the price-dividend ratio got. That was in the year 2000. It got almost up to 100, meaning that the dividend yield was not much over 1% on the S&P 500. On the far right, we have NIPA, National Income and Product Account CAPE, which Jeremy has proposed. And I'll come back, where we're using the earnings for the national accounts rather than for the S&P 500.

So here are the various ratios that [INAUDIBLE], [INAUDIBLE], and I looked at-- the CAPE ratio, dividend-price ratio, earnings-price ratio, book-value-to-price ratio, sales-to-price ratio, short rate, and VIX. How much do those things forecast? And we're looking at various horizon returns.

So here are regressions using the sample period shown at the right side of returns over various horizons. What we're showing is the t statistic and the r-squared. Now, it's 1 over CAPE as the only dependent variable. Then dividend/price ratio, earnings/price ratio, and book-to-price ratio. CAPE seems to do dominate pretty much at every interval.

I might add we should have maybe put a slide that in the third edition of my book, Irrational Exuberance. I added a regression of 10-year real return on CAPE and-- I'm sorry-- 10-year excess return between stocks and bonds-- and regress that on CAPE and a 10-year real interest rate with the whole sample back to 1881.

If you put two variables in, both CAPE and the interest rate, you get an r-squared of 0.41. We're getting toward halfway predicting the excess return. So if you look at that regression now, it's predicting a modest excess return of the stock market over the bond market of about 1% a year for the next 10 years.

Now, there's an important-- I've learned so many things from Jeremy. One of them-- he is the only man in the world who has made a campaign that argues that the stocks are safer than bonds if you look at a long horizon. And he's got the data to back that up. So I'm thinking, what about for a long-horizon investor? My CAPE excess return regression predicts 1% excess return and a lot of the volatility in the short run.

But over longer runs, maybe not. So what I'm thinking is that stocks are still a good investment-- maybe one-- if you were confined to the United States and equities and bonds, you might put most of your portfolio in it. I'm not sure I'd go that far. But simple Markowitz analysis would suggest that, assuming zero correlation between stocks and bonds.

But here are some other variables-- the sales-to-price ratio, the T-bill rate, and the volatility. You'll note that in every one of these, they underperform the CAPE ratio in predicting 10-yea or various horizon returns. We only go out five years on these charts.

So that's the bottom line about CAPE. We also [INAUDIBLE], [INAUDIBLE], and I also looked at other countries. And here's a list of countries. All but two of them, the CAPE ratio outperforms-- on the one-year horizon, outperforms in terms of higher significance. They're not that significant in the one-year. It has to be long horizon returns that are more forecastable.

So now, this gets now to Jeremy's. I think we might hear more from him about this in a minute. I don't know. But Jeremy thinks that we should replace the S&P earnings that I've been using in the CAPE ratio with national income and product account earnings. Now, that's an interesting idea. And it's something I would do. I haven't done it yet though because I have doubts.

So Jeremy thinks that maybe the national income and product accounts earnings are more-- have higher integrity because they haven't been changed as much in the procedure. We have the FASB, which is constantly announcing rules changes. So maybe the NIPA earnings are not as consistent through time as our S&P or the earnings series that I use.

I think that's an interesting point. He gives some evidence to suggest that might be right. But the problem is there are problems with NIPA earnings as well. And that's why I'm hesitating to go to the alternative suggested. NIPA earnings are part of the National Income and Product Accounts. So they don't treat foreign earnings. They try not to include foreign earnings.

They do not reduce-- they do not deduct bad debt losses. They are really designed as a breakdown of national income. And the changes that they've made are often changes that are well-intentioned. It's not like the FASB are bad guys here trying to confuse everybody. They are making changes to improve the integrity, maybe reacting to practices that they became aware of.

The other thing about NIPA earnings is they have a different trend. They have a-- going back to the beginning in 1929 when national income and products accounts began, NIPA earnings have tripled relative to S&P earnings. Why is that? It's because they're measuring all earnings, including not just publicly-traded stocks, but S corporations, which have increased in value or just other companies that were never in the index, but have been growing.

So Jeremy divides by the S&P divisor to remove the trend. And when you do that, you can see in this chart that the NIPA earnings seem-- there's a level difference that we've corrected for. But the NIPA earnings seem to go along with the reported earnings more or less. So he's gotten out the trend.

But the problem is-- I think-- I worry about the divisor as a detrender, because it wasn't designed for that. The divisor for the S&P 500 is designed to take account of the substitutions they made and the rights issues and the share repurchases that companies in the S&P 500 made. It doesn't necessarily match up to the whole economy.

But it does look like it did do a good job of detrending earnings. So maybe it's all right. Other alternatives to CAPE are operating earnings, cash flows. We estimated a CAPE based on those alternative earnings measures. And we got pretty much the same chart. Also, this is a point that we owe to Cliff Asness-- the earnings that we see in CAPE as we define it, are very much lowered by the recent experience. In fact, we're above the long-run trend of 10-year average earnings.

And finally, we did one more exercise, because Jeremy and other critics have argued that 10-year average-- what's that?

GILLIAN TETT: Almost out of time.

ROBERT SHILLER: This is my last slide. So I've timed it perfectly here almost. 2001 and 2008 were unusual years for earnings. And so critics say that we should have removed them from our calculation. But here we did remove them, and things are not a whole lot different. So it remains that CAPE is high.

So the bottom line is, Jeremy has raised some interesting points that we may someday use to adjust our procedures. But because we have these various doubts, we haven't done that yet and still stick with the view that the stock market in the United States is highly priced.

GILLIAN TETT: Well, thank you, that's--

[APPLAUSE]

So Professor Siegel, the floor is yours.

JEREMY SIEGEL: Thank you, Bob. I love that picture. I'm jealous you kept all your hair. I didn't keep mine. It may be a little grayer than that, but yes, we've had a wonderful 50-year relationship meeting in September of 1967 just before our first year at MIT.

So let me talk a little bit about my view. First of all, the long run-- stocks, bonds, treasury bills going all the way back to the beginning of the 19th century-- gold and the value of the dollar. And the trend line and the returns that you see on the last 6.7% for stocks, 3:5% for bonds-- wow, 3.5% for bonds is real-- 2.6% for bills, gold, 0.5%-- the dollar, this is basically minus inflation depreciating a 1.4% per year.

But we're talking here about valuation. Almost all of you have Bloomberg screens. So you can get this off your Bloomberg screen. The P-E ratio of the S&P over about 63 years. The median actually is up almost 17 over that period. And there it is. And we are above the median. You can definitely see that in the period when we had double-digit interest rates was the below-10 period. And Bob and I agreed, the year 2000-- all-time high-- a P-E of 30 or 21, 22 times last year's earnings.

I want to talk a little bit about earnings though-- definitions. There are really three types of earnings. There's firm reported earnings. That's what we get on CNBC-- the most liberal of all. They often exclude asset impairment, severance costs, et cetera, and so on. The last 12 months-- $114.19. This year's estimated $134.52.

A definition I like much better is S&P operating. It excludes a lot of the liberal-- it takes away all options that are given to employees-- option impairments except for financial, severance costs. And you can see it's lower. This is actually the earnings that Warren Buffett prefers to use. S&P use to call it the core earnings.

And then finally, there's GAAP earnings. Now, the important thing about GAAP earnings. And I am going to make this point later-- is they have changed their definition over time. In the late 1990s, they began to mandate firms, mark down assets, and include that in current earnings, whether the asset was sold or not.

And by the way, it's sort of asymmetrical, because if they go up, they make you sell it in most instances to prove that it's gone up. And as you can see, it is a lower number than the S&P operating earnings. What do P-E ratios mean? Well, they're important, because the inverse is the earnings yield. And over the last 140 years, the average P-E ratio has been about 15-- and 1 over 15 is, believe it or, not 6.7, which is the long-term real yield on bonds.

We're not going to be there now. When I did this May 15, S&P was 2,400, a little bit below that. We were selling 21 and 1/2 times last 12 months of earnings-- 18 and 1/2 times this year's estimate of operating earnings, which I think is my definition-- 1 over 18 is 5.4%. So the earnings yield associated with the current, if maintained at this level, will give you a 5.4% real.

This is about 5 percentage points above the bond rate-- what we call the equity premium. And by the way, that's greater than the 3 to 3.5% that's the long-term average. And by the way, I'm using the geometric averages over here on equity premiums. Even if we say a 20 P-E, if maintained, gives you a 5% real return.

And if the Fed reaches a 2% target, which we're very near, that's a 7% nominal return. This is still well above the margin of bonds. 10 year tips are 40 basis points. A 5 percentage point real return, again, is well above that. The reason Bob on his CAPE ration, and we're going to talk about that in a moment, is so much lower is because he assumes reversion to the mean CAPE ratio. And I'm going to talk about that in a moment.

By the way, the 5% real return on a 20 P-E today consists of a 2% dividend, a 2.5% buyback, and by the way, the buybacks are by far the major reason why the divisor changes over recent times at least-- and only 5/10 of a percent on capital expenditures. Growth is not that important, because it requires capital.

Here's world P-E ratios. You can get that off your Bloomberg screen also hitting [INAUDIBLE] WPE. You can see it for North America in the first segment. Europe is cheaper with a 15 average. Last year, it was a 13 average, and I said it was a good buy. At the bottom, we have Asia. Really well except for North America, we're very close to the 15 ratio. We don't have emerging markets here. But there are also in the 15 region.

Then we talk a little bit about the CAPE ratio. Bob defined what it was. I think his work with John Campbell was absolutely astounding back in the late 1990s-- 10 years of earnings. It was very much, again, above the long-run average. And I you predicted around a 2% real return for 10 years. By the way, I think the original r-square was around 1/3. If you say it may have gotten a little higher, this is astounding as a 10-year average. So it's very, very strong.

But if you look at the CAPE ratio, and I've shaded this area, you notice that over the last 25 years, except for a few months, at the very bottom of the financial crisis recession-- bear market. We have been well above the CAPE ratio. By the way, it should be a little bit higher than it is now, but still well below the 2000. And I began to wonder why that's the case.

This is Bob's real per share reported earnings. Bob uses GAAP earnings for his real per share. But what was interesting-- the last two recessions, the mildest in our history, which was 1991, and the financial crisis. As you can see, a tremendous decline in earnings. By the way, that was the Great Depression, not beforehand-- the biggest decline was World War I. I looked at the financial crisis. I said, just a minute. We had nearly twice the decline of earnings in the financial crisis.

GDP dropped by only 20%. That gave me a clue that something else was happening to the earnings. And in fact, it was in the early '90s where they began to mandate the write-down in the earnings. There is a difference. I am just going to show you S&P earnings. This is S&P operating earnings. You see it doesn't include the write-downs. You see how much milder it is in the last two recessions. And don't forget, Bob uses a 10-year average. So we're still in there. In a few years, we'll be out of that tremendous decline that we had there.

Also, NIPA profits is also higher. Operating earning-- it's even higher than the others. And Bob gave you some warnings on that. But again, it does not show that tremendous decline. And by the way, it is a non-weighted 10-year average. Am I right, Bob? So that 2008, 2009 tremendous dip is in equally with all the other numbers that you have in there.

So judging the CAPE ratio, I think it was the best predictor of real returns when it was first published in the late 1990s with John Campbell. It still has a lot of things to say for it. But I think it has been overly bearish in recent years. As I say, only nine months out of last 25 years has it been below the long run average. One, as I mentioned to you, the changed definition of GAAP earnings over time.

Secondly, I believe we are for a long period, although you know, we only have to wait, in a permanently lower interest rate environment. We, of course, are in slow productivity growth. We don't know all the reasons. We don't know when we're going to snap out. We know we're in slow population growth. That's going to be a long time.

I do think that changes the normal P-E ratio, although it's hard to see it from historical standards. And that's why I believe the P-E ratio should be higher. I'm not, again, believing we're going to have 6.7% into the future. I think it's going to be lower.

And the third reason, I think it's very important-- over the last 140 years or more that we've had the CAPE ratio. Think of how easy it is to own the market portfolio. Think of an investor in 1881 or 1923 that tried to diversify across hundreds of stocks in proportion to get the market and the transactions costs that were involved. That could have subtracted from 1 to 2 percentage points from his or her return. A 15 P-E ratio today is very different than a 15 P-E ratio in 1881.

They were actually probably only getting a 5% risk-adjusted return, which would be associated with a 20 P-E ratio. So I believe there should be a small upward trend to P-E ratios to take into account of the tremendous ease we can eliminate or reduce individual idiosyncratic risk by that and should 15 be right over time. This is my last slide. I know it's up.

Again, let me just say, I don't think 6.7. I think 5.0. And I think the low interest rate environment that we have been in and I think will be in for many years makes stocks still definitely the asset of choice for your portfolios. Thank you very much.

[APPLAUSE]

GILLIAN TETT: Well, as I said, we do have two veritable luminaries of the world of the 20th century and 21st century finance with us today presenting two distinctly different perspectives. I'd like to once again just quickly off the audience-- I'm curious about this as to where the audience stands with these two different perspectives on whether the stock market is overvalued or undervalued. Hands up in the audience today who thinks that the equity market is overvalued and who would lean towards the Professor Shiller's perspective on the markets.

OK. And hands up who tends to lean more towards Professor Siegel's view that actually stocks are not overvalued? Well, we have a fairly evenly split, I would say.

JEREMY SIEGEL: Yeah, that's good.

ROBERT SHILLER: By the way, I have questionnaire surveys of individual and institutional investors where I ask that question. And most people think it's overvalued-- maybe not the august people at present.

JEREMY SIEGEL: I think you also have to ask the question-- even I said it's overvalued relative to its historic norm. Is it overvalued relative to the returns that are available in other asset classes? And that's a different question.

ROBERT SHILLER: We agree on that point. I just got through saying that you might hold most of your US portfolio in stock, because 1% excess return is an excess return. And you said-- have you taken back your claim that stocks are less risky than bonds for the long term?

JEREMY SIEGEL: Well, first of, when I first wrote Stocks for the Long Run, there were no TIPS. You had inflation risk. So it did turn out that over 30-year periods, the variance of real stock returns was that. Now that we have TIPS, clearly you have an ability to reduce them. But look at the yield. The 10-year TIPS is 40 basis points. A 20 P-E stock ratio was 5% if it maintains at that level.

That is a huge differential. I do not think it is going to permanently move back to 15. Temporarily in a sell off, anything can happen. But I don't think 15 is going to be for quite a while the so-called right or normal P-E ratio.

GILLIAN TETT: Right. Professor Shiller, are you working on the assumption, like Professor Siegel, that interest rates are going to stay low for a long time and inflation will stay low for a long time?

ROBERT SHILLER: This is a deep question. It's a question about secular stagnation. And it's about new normal. I don't think that anyone has-- there's no received wisdom about what causes it. But the fact that long rates are so low suggests that there's a general feeling that we are in for a long period of secular stagnation.

Now, I was president of the American Economic Association last year, and I gave my presidential address entitled "Narrative Economics." There is a secular stagnation narrative. This is not science. It's a narrative. Nobody can prove it right or wrong. But people believe it now. The last time we had this narrative was in the Great Depression. And we were shocked out of it by World War II.

After World War II, people thought the Great Depression is going to return. That was the standard view among economists. It did not. It shows to me that public opinions on things like this are changeable. And I don't know that we'll stay in this secular stagnation narrative for a long time.

GILLIAN TETT: Well, if any of you in the audience have not read Professor Shiller's wonderful address to the American Economic Association, I really would urge you all to read it, because I'm deeply biased. I'm trained as a cultural anthropologist. And the address actually tries to weave together some elements of anthropology and economics. But you made the point that economics is not a science-- dare I say it. Ideas change. It's embedded in cultural context.

JEREMY SIEGEL: I would just want to mention-- remember, growth is two parts. One is productivity, which is often the secular stagnation part of it. It is about two-thirds of growth historically. The other third is population growth. And at least we know demographics-- birth rates could go up. But certainly, I think we're at the slowest growth of US population certainly working for us than we've almost ever been.

ROBERT SHILLER: Except in the Great Depression-- it was also very low.

JEREMY SIEGEL: Well, but our real incomes-- real GDP fell 25% in the Great Depression.

ROBERT SHILLER: That was different.

JEREMY SIEGEL: And we're well above our lows. And we were at all-time highs. We were talking last night--

ROBERT SHILLER: We're just continuing an economics discussion.

GILLIAN TETT: OK, well, for the benefit of everyone else in the room who wasn't there last night's-- can I ask though about the population question? Robots could change that. With robots around replacing human workers, do you need to have the same level of population growth to create productivity increases and growing economy as before?

JEREMY SIEGEL: Well, it doesn't go all the way--

GILLIAN TETT: Demographic growth, sorry.

JEREMY SIEGEL: You want to go Bob first? You want to go that way.

GILLIAN TETT: Either of you, yes.

JEREMY SIEGEL: Technological change has been part of our life since the Industrial Revolution, replacing workers with more efficient machinery is-- I regard robots ads is it going to accelerate? There's talk about that. That would accelerate it. And by the way, that would bring up real interest rates and probably bring down stock prices, because you do have a capitalization ratio there-- also increased real economic growth, which could partly offset that on the other side.

But there's a lot of debate on that issue about whether we are on the verge of another acceleration in productivity. What do you think?

ROBERT SHILLER: I think we should regard-- first of all, it's the narrative. Everybody is talking about robots. That's what we know for sure. What is the reality of this? We don't know. It's saying that-- I know the Luddites were 200 years ago complaining about automation. And they were wrong. That doesn't mean it's not coming now.

And what I'm thinking-- you must be aware that robotization is accelerating. You can talk to them. They kept saying Alexa is getting smarter. OK? She's learning new tricks. She knows more than most people already. And she's a machine. These things are accelerating now. I think that there is a serious risk that our society will be fundamentally transformed in coming decades in a way-- and by the way, also, during the Great Depression, this fear was also prominent. They were talking about robots then. And there was a lot of narrative then about automation.

And they thought that millions of these jobs were lost. We forget this now. But the average person thought that a good part of the depression was what they called technological unemployment. That's their term.

So those things are still scaring us. And I think that we should consider now an action plan what to do about robots. Bill Gates wants a robust tax. He gets ridiculed for that. But I'm one of his supporters. Or we should develop a sort of social insurance program now, I call it inequality insurance, that will deal with the robot revolution if it turns out to be as bad as people are afraid of now.

GILLIAN TETT: Right, well, those are big, big questions, as you say. We have a large second question from the audience about very tangible investment themes. So I'm going to run through some of these questions to go back to your difference in perspective and approach. We have a question for both of you. "What are your expectations for the real bond return?"

JEREMY SIEGEL: Well I mean we have it certainly in the TIPS market. I think it's just under 1% for 30 years. There may be a little more risk there. The 10 year is kind of the benchmark-- 40 basis points this morning. I do think growth will go up somewhat. So 5 or 10 years from now, I think it will be somewhat higher. But I don't think-- when TIPS were floated in 1997, the first TIPS, 10 year. It was a 10-year TIPS. It went off at 3.4%. Interestingly enough, that was exactly the GDP growth in the post-war period up till that time. It matched the GDP growth. It's actually gone down a lot more because of some other reasons. And the GDP has gone down. But there was that magic. GDP growth, if we get it up to 3%, you're going to have a rise in the TIPS yield-- no question about that.

GILLIAN TETT: Professor Shiller?

ROBERT SHILLER: I think there's an embarrassing lack of understanding about interest rate trends. Interest rates and inflation have been trending down since the early 1980s. I'm thinking that we have to understand the public narratives that underlie this. I think it has been gradually changing. I can't pin down these things. But there was a cost push narrative. There was an inflation is inevitable-- it's always going to get worse narrative, which has faded.

Now where it's going-- It's been declining. As Jeremy says, real rates have been declining. It has nothing to do with the financial-- or little to do with the financial crisis. They've been trending down ever since the mid '80s.

JEREMY SIEGEL: And by the way, I don't think it has that much to do with central banks. I know I'm sort of on left field there. I think the main thing has to do with risk aversion, but slow growth, which influences real rates-- tremendous demands for liquidity both by the governments and by the financial institutions and by regulation. We all got a letter. Our money market fund could no longer serve as collateral against borrowed stock. It had to be government securities.

GILLIAN TETT: But of course, very few of those elements are actually in the financial models that people are using to try and project what's going to happen. That's a big challenge for anybody who's trying to make sense right now that, as Professor Shiller pointed out in his speech, that much of this is not in the current thinking of economists. We both agree to agree to find anthropologist or a psychologist [INAUDIBLE].

ROBERT SHILLER: As former president of the AEI, I have to stick up for economists. But they do miss things. They missed whole financial crisis as an example. And they would have done better if they were more focused on what people are saying-- what they're thinking.

JEREMY SIEGEL: Well, they actually would-- believe it or not-- they would be more focused if they had re-read The General Theory of Keynes. I mean it was over-speculation there in the stock market-- this was in the real estate market and in the shadow banking system-- a run on that. It crashed the financial institutions.

And in my opinion, the only reason it didn't get anywhere near as bad is the liquidity and the lessons that Bernanke had learned and we had learned as economists. And I really say that-- and Bob is saying sticking up for economists. The lessons we learned from the 1930s-- and I say thanks to Milton Friedman's monetary history-- say don't let the money supply crash as it did in the '30s. And that's the reason we had 5.8% decline and not 25% decline in GDP.

GILLIAN TETT: Well, we have another very interesting philosophical question here-- actually quite practical question. "How are stock valuations affected by the huge public and private debt overhang in the world?"

JEREMY SIEGEL: OK, so-- that's a hard question. We know our-- is it is it private debt we're worried about or government debt? We know in the long run without changes, our entitlement systems are not sustainable as they stand today.

And most of it is Medicare. There just will have to be changes. It doesn't have to be done today or tomorrow. If you actually look at the CBO and look at their government debt-to-GDP ratio, it's actually a very mild rise. And then around 20, 30 or so, it begins to really get extremely cheap.

But those are the major, I think, overhangs. The Medicare is an unfunded one. The Social Security, we have more "funding" on it. I worry-- that is more important to me than the private debt. I do not think we are overly-leveraged Americans on private debt today given the income levels.

ROBERT SHILLER: I might just refer to my New York Times article on Sunday in the business section, where I talked about the origins of the speculative attitudes that seem to lie as causes of the boom in real estate preceding the financial crisis. Now, I don't claim to have this all figured out. But I think there was a change in narrative over the years that emphasized risk taking-- borrowing and risk taking and a speculative culture, which ingrained everyone's thinking, especially people who hadn't succeeded in life. They thought they could flip houses now as a technique.

So the kind of thing that I would look at to understand the change mentality-- one of the examples I quote in my op-ed piece was from a book written in 2004 called How to Flip Houses. And in this book, aiming to sell people on the idea of doing that, the author defines what he calls the OPM principle. You know what the OPM principle is? It's Other People's Money.

And what he says is, as much as you can, bet other people's money as much as you can, as if that's a smart thing to do. And I looked through the book to find out where does he say that's leverage, and there's a downside to leverage. It isn't there. These are just examples. It's hard to be systematic.

JEREMY SIEGEL: Are you saying real estate is overvalued now, Bob?

ROBERT SHILLER: Real estate, according to the S&P Case-Shiller indices, which I've [INAUDIBLE]-- real estate rose 75% in real value from 1997 to 2005. And then it fell by-- it fell to only 12% above the 1997 level. It essentially went all the way back down-- almost all the way back down.

And now, it's halfway back up. So it's nowhere near 2005. But what I'm seeing now is some clues of a spec-- it's not intense yet. It's not bubble-like yet. But there are people who are getting excited about flipping houses. And I counted the number of books currently on Amazon on how to flip a house. And it was between 300 and 400.

And I looked at the books, and I thought they sounded optimal-- I didn't look at all of them. I [INAUDIBLE] it. It's coming-- at the moment, it's trending back. So there was a complacency. There was a feeling that home prices will never fall. They didn't ever say that. It's hard to find someone saying home prices never fall. They just don't mention the possibility. Now we've seen them fall, and we're chastised. So we're not in that bubble mentality now. That same bubble mentality affected mortgage lenders. It's just the home buyers. But we're not there yet.

GILLIAN TETT: So Professor Siegel, there's a fairly tactful question here-- "Is it fair to compare earnings to bond yields given that investors have no control over how earnings are reinvested?"

JEREMY SIEGEL: Well, theoretically-- so let's just take a 20 P-E ratio, which again, we're about at now. That's a 5% earnings yield. They could get it all back as a dividend 5% yield. They're giving 2% back. They're buying back the other 3%. They're buying back about 2.5% of their shares. This is S&P. The remainder 0.5% is CapEx. There's very little CapEx.

As you know, this is one of the complaints about this expansion is the fact that firms are satisfying demand, and there isn't the need for capital. Theoretically, and I have a chapter in my book, Stocks for the Long Run, that shows it doesn't matter. They do dividends. Or they go do buybacks. Or they do it on growth.

Theoretically, that should still yield a 5% yield to investors at that maintained ratio.

ROBERT SHILLER: We haven't mentioned Donald Trump.

GILLIAN TETT: Well, I could spend the whole hour talking about Donald Trump.

JEREMY SIEGEL: That [INAUDIBLE] lesson.

GILLIAN TETT: But please give me your views on Donald Trump and the CAPE theory-- the CAPE ratio.

ROBERT SHILLER: The big question now is what changes we're seeing. And there is this Trump enthusiasm that did boost the market after the election. That might put me more in your camp, except I don't trust it.

JEREMY SIEGEL: Well, you know-- some of you-- I don't know if some of you watch Barrons. I was interviewed on CNBC last week. And I was feeling very controversial. So this was when the Dow was down 350 points. And I said, well, he's in so much trouble if he got on the office and said, I don't like this job. I'm resigning, I'm giving it to Michael Pence to be president, the Dow would go up 1,000 points.

Boy, did I get a lot of comments on that. But let me say the source of that-- the source of that is that the market likes the Republican agenda. It's not that they like the Trump agenda. Now, there's intersections between the two that are the same. But those that are different, the market doesn't like-- the trade restrictions, the barriers, and all, H-1B visas, and all that. That's not what the market likes.

So in a way, Trump became a way that the Republican agenda could be adopted into law. And as I mentioned, if he stays unpopular, and if the voters take revenge against the Republicans and turn the House Democratic in two years, hey, a lot can be done in two years. When Obama took over after the financial crisis and controlled all three houses-- executive, and House, and Senate, Obamacare was passed. Graham-Dodd was passed. There were thousands of other regulations that were passed. You can do it. You could get corporate tax reform. You could get the type of reforms. And unless all three go the other way, you won't get it undone either.

So my feeling is that it's the Republican agenda that sent the market up and that supports the market today. And I believe that a strengthening of the Republicans and a weakening of Trump-- I think Trump will sign what the Republicans finally come up with on corporate, personal, and regulatory reform. And that can be done certainly before January of 2020.

GILLIAN TETT: Well, going back to Professor Shiller, certainly, the narrative has changed a lot, hasn't it?

ROBERT SHILLER: Well, we have an unfortunate division in this country. And there are different narratives, depending on where you live-- red state or blue state. And we have among a substantial element of the population who think that Trump is a revolution long overdue. And they are very motivated on it. If Trump were to be impeached, I don't know whether you included that, that seems to me such an intense thing that I think it would hurt the stock market.

JEREMY SIEGEL: Well, Clinton was impeached, and the stock market kept on going up.

ROBERT SHILLER: I'm thinking of during the 2011 debt crisis, markets were very volatile at that time. I don't know what will happen. It seems to me that I would not welcome a contentious-- I'm hoping Trump does well in his position.

GILLIAN TETT: Professor Siegel, another question-- you discussed your S&P top 10 dividend strategy in your book, which is beating the market. Do you expect the strategy to continue to work-- if people still need-- do people still need to buy your book?

JEREMY SIEGEL: Well, first of all, it's difficult. Under Obama, taxes on dividends went up. There's still no question that capital gains have a tax advantage, because they can be deferred. And that affects the way you look at valuations. Bob did test the dividend yield. I take the earnings yield, because now, I think we've never had as many, in the last 10 years, buybacks, again motivated, not just by the tax law-- also options that are given out.

So in a way, I'm really a value investor. In a way, Bob and I agree. When P-E ratios are high, you're going to have lower-than-normal returns. I just think the returns right now are still very generous on equities, especially given the alternatives. Real estate has gone up, and of course, bonds have gone way up to give you almost a zero real return long term.

GILLIAN TETT: And another question is, do your analytics still apply to non-US stocks?

JEREMY SIEGEL: To what?

GILLIAN TETT: To non-US stocks.

JEREMY SIEGEL: Oh, yes, which are cheaper or were much cheaper. I will admit earlier I was saying, a year and a half ago, buy emerging markets, wow! They have 12 P-E growing two to three times as fast as the US. Yeah, I was way too early, and everyone knows that's a common factor. But they're still below. You saw a 15 P-E in Europe versus 1819 in the US, depending on what earnings-- 15 P-E in emerging markets. Half the equity in the world is US. I definitely think you should not ignore global diversification.

GILLIAN TETT: Right. Well, we have a couple of questions for Professor Shiller. "Does all the analytics of the stock market in the US being overvalued together with Professor Siegel's analysis indicate that bonds are going to outperform in the next 10 years?"

JEREMY SIEGEL: They could all perform, absolutely.

ROBERT SHILLER: Could. They could. is that your median estimate?

JEREMY SIEGEL: No, I said, and I predicted excess return is about 1% for the next 10 years. So that means if you add inflation into that, that's talking about a 3% stock market return-- not bad--

ROBERT SHILLER: 10 years too [INAUDIBLE].

JEREMY SIEGEL: But that's, again, that reversion of P-E down. That's a major factor that gives you the pessimism.

ROBERT SHILLER: See, now, more thing that-- you say you're a value investor. But you haven't mentioned, or I didn't hear you, talk about why-- you said something about dividends-- capital gains are preferable to dividends, because they're deferable. I think that's a minor part of it. A big part of it is that some stocks are talked about. And they're exciting to people, and they bid the price up.

Now, you would say then maybe short sales strategies would offset that. But in fact, in history, they don't seem to--

JEREMY SIEGEL: There's always going to be a few stocks are going to be outsized, and we can't explain the P-E.

ROBERT SHILLER: And you want to avoid them. In fact, I remember you writing in 1999 about the price-earnings ratio. And I don't know exactly how you put it then, but it sounded like it was behavioral-- like people are too excited.

JEREMY SIEGEL: Well, I would say back then, it was called big cap stocks are a suckers bet. People say, Jeremy, you're never bearish. I say, no, look at that 1999 op-ed piece on the Wall Street Journal. I said, the S&P, by the way, the tech sector the S&P-- of the S&P, which only added stocks that were earnings, was 80 to 90 at the peak of the bubble. What is it today? 12, 14? I'm not going to try to explain Amazon, or Tesla, or those.

ROBERT SHILLER: What was happening then was analagous-- the same thing that happened in the Gold Rush in 1849. What was historic? This is the time to stake your claim. Don't worry about whether they're actually-- you've found gold out there. You've got to buy it before there are earnings. This is the internet bubble. The idea that earnings don't matter anymore. So people had a narrative, which justified buying at crazy price-earnings ratios.

GILLIAN TETT: Well, we could carry on discussing this for at least another hour. We haven't reached a resolution. I'm sure you never will. The audience is probably split fairly evenly on either side. The one thing I can conclude, which is crystal clear to me, that if you want a new money-making venture that will have a lot of value, you should write a book together. Or you should have essays together going back and forth, because the other thing that's crystal clear to me is you don't actually need a moderator. You are happy asking questions of each other. Carry on, please, and thank you.

[APPLAUSE]

ROBERT SHILLER: Thank you.

ANNOUNCER: Copyright 2017, CFA Institute, all rights reserved. This program is designed to give accurate and authoritative information in regards to the subject matter covered. It is distributed with the understanding that CFA Institute is not engaged in rendering legal, accounting, tax, investment, or other expert advice. If legal advice or other expert assistance is required, the services of a competent professional should be sought.