[MUSIC] Hi there, in this third video we're going to talk about another bubble, which was due to excessive valuation. And that's the so-called dotcom bubble, also known as the TMT bubble, TMT standing for technology, media, and telecom. So as a first chart, which I'm going to show you, this one, there's going to be a quiz for you. And you see that this quiz is going to be similar to the ones you had to do for the 1987 crash. Basically I would like you to find out by how much the equity market is overvalued relative to the bond market using the Fed model. So looking at this first chart, you see it here, the evolution of the equity market between 1997 and 1999. That's the green line for equities and the same for the bond markets. And well we know by now that it's not enough to just compare where the equity market is relative to the bond market and say there's so much overvaluation in this instance just comparing the two price indices, 81%. You need to compare two valuation yardsticks. And we know from the Fed models that these valuation yardsticks is the earnings yield for the equity market and the bond yield for the bond market. So question to you, quiz to you, if we look at this chart now and we see that the bond yields is at 6.7%, and the earnings yield is at 4.2%. Using this 4.2% now, what would be, what should be the P1, the equilibrium price of the stock market such that the earnings yield goes from 4.2 to 6.7? Okay, so here's the solution to the quiz. We start from the earnings yield at 4.2. And we want to find out which P will give an earnings yield of 6.7. So we rewrite the equation here, 4.2 times P0 divided by P1 is 6.7, and we find P1. And we find that P0 relative to P1 is that 6.7 divided by 4.2, so today's stock market price is 59% above what it should be according to the Fed model. So there's an overvaluation of where the stock market should be using this model of close to 60%, which is huge. And did you remember when we discussed the biases which may occur when we want to buy a security and we have difficulties in selling that security when it's making losses? And we used the example of Nokia. Here we'll show you here the same chart, and indeed it was a bubble. And we identified this bubble by some kind of similar yardsticks. Here it's not the earnings yield, but it's the inverse of that earnings yield. It's the PE ratio. And we use the PE ratio of Nokia, of the telecom sector, and possibly of the whole market to find out whether there's an excessive valuation on the equity market. Indeed, with what happened in 2000, the market peaked, the NASDAQ index was at more than 5,000, and it crashed. And here we see that peak-to-trough the equity market tumbled by 46%. This is a very painful bear market. It lasts us two years. And I remember I was working in a bank at that time, in an asset management company. And every day you walk in and you see the market goes slightly up, and then down, and then down again the next day, and then down again the next day, little by little. And you see all these fund managers walking in their office and being really grim and really sad. And what happens in the market, basically, impacts your personal life and makes it very difficult to stay motivated. So here you see the two examples. 1987 crash, boom, you get a sharp drop in the equity market. And then it stabilizes and resumes pretty swiftly. Here we get a painful bear market because it lasts two years. It's more abrupt, the fall is more than 1987. But what is more important, it lasts much longer. So it impacts people behavior and happiness [LAUGH] to a much larger extent than the 1987 crash. Okay, so the Fed model is a useful tool to assess whether equities are over or undervalued vis-a-vis bonds, but the model also has its limitations. And they are listed here. Basically, the model is a simplification of the real world on 3 grounds, I would say. The first one is it assumes when we talk about earnings yield, it assumes that basically looking at the earnings of a company, this is what you would gain. So this assumption implicit there is that a company generates earnings, and they give all these earnings to their shareholders. So we basically here assuming that the dividend payout or ratio is 100%, so that's assumption number one. Further, it assumes, and more importantly, that there's no risk premium for holding risky assets. Basically you say you hold equities or you hold bonds, and basically you're comparing one against the other. But normally you should be rewarded with a risk premium for holding equities vis-a-vis bonds, and here the Fed model does not make this assumption. It further assumes, the Fed model, that basically we're comparing the earnings yield, and this is defined in real terms. It's E divided by P, and we are comparing that to a nominal measure of income, the stream of income for the bond market which is its yield, so the bond yield. So here we have a discrepancy. On one hand we are saying the stream of income for the equity market is in real terms and for the bond market, it's in nominal terms. So it's a discrepancy here. So now we know that equities can be overvalued vis-a-vis bonds. So i.e., you can have a bubble. But identifying a bubble does not tell you when it's going to burst. And indeed, I remember early in 1999, I used the Fed model. I was a Chief Investment Officer in a bank here in Geneva, and I used that model. And I waved the red flag and say, equities are too expensive vis-a-vis bonds. But the problem is if you go underweight equities versus bonds when you identify the valuation excess, you may miss the last leg of the rally in equities. And that's where you make a lot of money. So the key question is, how do you invest your money in an overvalued market? And the best way forward for me when you are investing in overvalued market is to stay invested just to capture this last leg of the rally, which typically occurs at the end of an overvalued market. But you buy some downside protections. And you stay invested, but you have some downside protections against a fall, which you don't know when it's going to happen. Okay, in conclusion of these three videos, we saw that bubbles may occur. And they are basically a speculative mania, which drives the price of an asset, typically equities, the risky asset, above, but it can also be bonds actually, above its fundamental value. Bubbles are hard to identify. Indeed, when we talked about central banks, this is the main argument that they use to say bubbles are hard to identify. Alan Greenspan, as we mentioned, used to say that we know that there's a bubble when it bursts. So that is an explanation for why they refrain to act preemptively. Although I think it would be a good idea, because we know that central banks are much more effective in preventing the formation of a bubble than in dealing with it once it's burst. The Fed Model is something which you can use to identify possibly overvaluation of a risky asset versus the non-risky asset, ie, equities versus bonds. [MUSIC]