In this special episode we review the relationship between war and financial markets. War is a tragedy. We are not minimizing the humanitarian tragedy of what is happening in Ukraine by focusing on the potential impact on financial markets. But we are offering a Rational Reminder for investors in a stressful time. Wars and financial markets have coexisted, and often been intertwined, for hundreds of years. Countries that have lost major wars have had their financial markets decimated, while global markets have been relatively resilient, even to major conflicts. In additional to the historical perspective, we offer some timeless lessons for investors to remember in times of stress.

Read the Transcript:

Ben Felix: This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision making from two Canadians. We're hosted by me, Benjamin Felix, and Cameron Passmore, portfolio managers at PWL Capital.

Welcome to a special of the Rational Reminder Podcast. We thought it was a good idea to do a special release episode kind of outside of our normal cadence just to talk a little bit about, not talk about, but address the situation in Ukraine, and we're not giving any political commentary or anything like that, that's not area of expertise, but we thought that what we could do is, as the name of our podcast suggests, provide a bit of a rational reminder in the type of empirical data that we usually speak to help people process from at least one perspective the things that are going on. And we also want to be clear upfront that war is clearly a humanitarian tragedy first and foremost and we don't want to minimize that by talking about the potential financial market impact, but financial markets are what we talk about, so that's the perspective that we're going to provide in this episode.

Cameron Passmore: And some people have reached out to us to ask us what sort of research is there around this to help them make decisions. So we're seeking to help answer that question. And for the record, we are recording this on Saturday the 26th of February, so things are very dynamic and fluid, but to give you a sense of the timing.

Ben Felix: The idea here is that we want to talk a little bit about the data with respect to financial markets in historical periods of war and political crises and things like that. It's the idea of people panic because this time always feels different. So we just want to give context for how different is this time from the perspective of financial markets, of course, not speaking to the human tragedy, which is certainly different.

So wars and financial markets have coexisted and often been intertwined for hundreds of years. I mean, the two go back together for a very, very long time, documented for a very long time, probably even longer than what we have documented. Countries that have lost major wars have had their financial markets completely decimated, but global markets, even when those periods have happened, for losing countries, global markets have been relatively resilient even to major conflicts. We'll talk a little bit about how resilient in the data in a minute. Despite the relative resilience of global markets compared to individual country markets, wars do tend to reduce returns to global stocks and also increase volatility. So we've seen a bit of that. I mean, Cameron, you mentioned the date, we've seen some decreases in prices, but we've also seen some almost paradoxical increases in market indexes in the last couple of days.

Cameron Passmore: Shocking market movements. Thursday was-

Ben Felix: Yeah, well, that's the volatility, I guess.

Cameron Passmore: But Thursday was one of the greatest turnarounds I think ever in some different indices, and then, again, Friday was a very dramatic up day, which I'm not sure many people had that forecasted on Wednesday.

Ben Felix: Some of the most extreme historical stock and bond market returns, both positive and negative, and some of the most extreme, but not actually the most extreme, we'll talk more about that in a bit, but they've been connected to or concentrated around wars, major wars, but I think that drawing predictive conclusions from that thinking, "Oh, there's a war or there's a major conflict that could get worse, therefore stocks are going to drop." I mean, we saw a single day example that you just mentioned, Cameron, but even in the aggregate data, that's not necessarily true. Some of the research that we'll talk about shows that between the First World War and the Second World War, people thought that they had learned lessons from the First World War and tried to reallocate their portfolios.

The First World War is quite interesting from a financial markets perspective, there was a big liquidity crisis when the war began probably because people didn't know it was going to happen. Neil Ferguson has a paper on this and he looks at a bunch of data and historical accounts. Neil Ferguson's a historian, although he sounds like an economist, he kind of has both perspectives. But World War I was a surprise. Based on the historical accounts and the financial market data that he had in the paper, he basically shows World War I, nobody expected it to happen. When it did, markets dropped, World War I was rough for financial markets. And then when World War II happened, people tried to reposition their portfolios but ended up losing out because stocks in the UK and the US actually appreciated during World War II for a local UK investor, for US investor in UK stocks, World War II was actually... We'll get more into that data in a second.

Why do people worry about situations like this? While there are, like I mentioned, some really extreme events that center around wars. The two most extreme cases, at least from the perspective of market returns, is the Russian Revolution in 1917 where investors lost everything, close to everything. I mean, Russian bonds continued to trade after the revolution, but steadily decreased in value. So like in the Dimson, Marsh, Staunton data, they zero it out after 1917, and likewise in the Chinese Civil War, which ended in 1949, in both of those cases, the Russian Revolution and the Chinese Civil War, investor's assets were expropriated, so safe to assume 100% losses for investors in those countries. That is genuinely scary.

Very interesting point that I came across while looking through that data is that if you had been investing in the company's listed on the St. Petersburg Stock Exchange from 1865 to 1917, so prior to the Russian Revolution, you substantially outperformed having invested in the stocks on the New York Stock Exchange. So from 1865 to 1917, you had about twice as much ending wealth investing in the St. Petersburg Stock Exchange as the NYSC. Of course, that changed in the snap of a finger in 1917 when assets were expropriated. So I look at that and think about the historical US experience, and we've talked about this many times on the podcast, but people look at the US market and think it's got the best historical returns, therefore that's where you should invest. But an example like that shows who knows if, of course it's a big if, but if the Russian Revolution had not happened and their stock exchange had continued on the pace it was going, maybe we wouldn't be looking at the US.

And the US hasn't had... Like the examples that we're going to talk about: countries on the losing end of wars, countries with revolutions like China and the civil war, revolution in Russia and the civil war in China where assets go to zero. The US has not had any cases like that, but that doesn't mean that it couldn't, it's an important historical lesson, I think.

So German stocks during and after World War I. So I looked at the period 1914 to 1922, German stocks lost more than 90% of their real value measured in US dollars during the war, so a few years after they really got crushed, but during the war German stocks dropped by 67% cumulative over the period. US and UK investors also lost in World War I, not as bad as Germany. US Stock investors lost 18% in real terms, investors in the UK lost 17% measured in USD, and that was buoyed by an appreciation in the pound relative to the dollar. A local UK investor in World War I lost 36% in real terms. Japan had a big increase in the value of their stock market during World War I. 1914 to 1918 they gained a real 63% cumulative in US dollars and the world index over that 1914 to 18 period lost 31% real in US dollars, so pretty bad.

Germany was hit again with big losses during and after World War II. The stock market dropped by more than 90% from 1939 to 1947, Japan was even worse, their stock market lost nearly 99% of its real USD value during and after World War II substantial, pretty crazy. And, again, like you think of the human tragedy in both of those cases, it feels bad just talking about the financial market aspect, both of those cases. It feels bad just talking about the financial market aspect, but that's what we're talking about.

Investors in US and UK stocks, this is the point I mentioned earlier, who tried to learn from World War I by getting out of stocks before the war, because I guess World War II was maybe easier to spot them World War I, at least based on what I read from the Neil's Ferguson. So investors tried to reposition, ended up losing out on positive real returns, with the US delivering 22% real and the UK 34% real cumulative in their respective currencies during World War II. Measured in USD, UK stocks actually lost over that period. But local US and UK investors had positive real returns through the course of World War II and after. A global equity investor through World War II lost about 15% real measured in USD.

Now, exposure to multiple risk premiums paid off a lot over this period. I didn't have these data for World War I, but for World War II, US small cap value stocks beat the US market by an annualized, so this is per year, rather the other numbers I was talking about were cumulative. This is per year 12% from 1939 to 1947. And US value stocks beat the US market by 6% per year over the same period. Those are some pretty big premiums there.

In the paper I mentioned from Neil Ferguson, got a quote from him here. He says, "In short, there is no simple recurrent pattern. Investors did try to learn from history in the late 1930s, but they mainly learned how to make new mistakes since the lessons of the previous war approved to have only limited relevance to the next one."

I thought that was pretty powerful, but it's also not really new. It's in the context of war, it's different, I guess. A different way to think about this situation, but financial markets are inherently uncertain, random and hard to learn from. We know that. Why would it be different from wars? Well, it's not.

Now, on average though, you can't use knowledge of war to predict which assets are going to do well. But on average, it is true that there is a decline in markets during times of war or crisis, which makes sense. We know stock prices theoretically reflect the present value of expected cash flows discounted at a rate that is reflective of risk. If times of war decrease expected cash flows or increase risk, which are both more than plausible during times of war, drops in prices are to be expected.

Now, what's the other side of that coin? Well, expected returns increase. If the discount rate goes through the roof and therefore prices fall, that also means the discount rate goes up and expected returns also go up. And again, it feels bad talking about this, but like we mentioned, this is what we're here talk about. It just feels weird to talk about, "Oh, look at the returns you could have had investing in Japan after World War II."

Anyway, in the years following the Second World War, so just looking at 1949 to '59, the German stock market increased at a rate of 61% per year. Adjusted for inflation in USD, that's big and that's real in USD. An investor who got into the German market in 1939 had lost nearly everything by 1947. But if they'd stayed invested and maintained the ownership of their property, which I'm sure was messy or at least interesting, they nearly tripled their initial investment by 1952.

The Japanese stocks appreciated at 28% per year from 1949 to '59. But the losses in Japan, remember they lost 99% of the value of the stock market in that case, they were severe enough that it took another 10 years, until 1969, just to recover the real purchasing power of a dollar that had been invested in 1939. That was bad. The German case ended up for investors being not so bad, but Japan, a little tougher.

Now, to contrast that experience of Japanese and German investors, a globally diversified investor, had that been per practical at the time, which it wasn't, but if you had a VT global market index fund, you lost about 15% in real terms, measured in USD throughout the course of the Second World War. Not nearly as volatile as the other cases, and you matched the ending wealth. By the end of 1959, the world market in Germany looked pretty similar in terms of their total returns.

Those are extreme events, deep in the left hails of stock market history and of human history as well. But I think it's also useful to take a broader perspective, because that's so far is kind of anecdotal with just a couple of major examples. 2006 paper, War, Peace, and Stock Markets, they looked at 440 international political crises over the period of 1918 to 2002. They got the crises from the International Crisis Behavior Database, which is from Duke University. It's quite interesting. It's all online. Based on those data, on average, an international political crisis starts almost once every two months. Not generally as extreme as what we're seeing now, obviously, but crises are not infrequent. The authors of this paper, they find that international crises reduce world stock market returns by approximately 4% per year. That's just in the returns. You look at historical returns, well, they're 4% lower per year because of international crises. They find large negative stock market reactions from world markets in the first month of a crisis, followed by lower than average returns during the remaining months and a partial recovery when they end. The reaction from stock prices that they find in this paper is stronger when a crisis involves basic values, like a territorial threat, a threat of grave damage or a threat to existence. And reactions are also stronger when a superpower is involved on both sides of the conflict and when the conflict starts with violence. You can see a lot of those traits in what we're seeing now.

Confirming what we've seen anecdotally at the extremes, like we talked about the World War I and II examples. This paper shows that investors in stock markets of countries involved in an international crisis see their markets drop about 2%. The countries involved see their markets drop about 2% when the crisis starts, and then additional 1% per month as long as the crisis lasts. On average, of course, in the full sample.

In the case of countries involved in the crisis, this was very important, those losses are only partially recovered when the crisis ends on average. Again, weirdness of talking about this stuff from a financial markets perspective. I've seen some people in the Russian money community, actually somebody made a post of, "Look at the relative price of Russian stocks right now, what a deal." And it's like, that feels kind of gross to talk about, but I actually responded with this quote from this paper that, "Listen, that's fine. But historically, if you look at past crises, the countries that are involved don't on average fully recover the initial losses that they sustain."

So I don't know if it makes sense to look at the Russian stock market and say, "What a deal." I don't know if it makes sense to do that morally, but also from a pure expected returns perspective.

Cameron Passmore: You wonder if globalization has a dramatic impact on this data going forward?

Ben Felix: Yeah. Globalization, but also globalization of the financial markets and easier access to the world index. Maybe the historical diversification benefits that we're talking about and will continue to talk about, maybe they're overstated because markets were not integrated historically and they're more integrated now. So it's a good question and it's an open question.

At the same time, you look at what's happened to the Russian stock market recently, then you look what happened to the world market over the same period. Well, that looks a whole lot like the data we're talking about right now.

International crisis have a strong impact, both on average returns and on volatility, with volatility increasing by slightly more than a third compared to its average at the onset of a crisis, and then decreasing by slightly less than a third when the crisis ends. The authors of the paper suggest that political uncertainty may help explain the volatility puzzle. That's a famous 1989 paper by Schwart. I don't know if I said that name well. But they posed this question of why does the volatility of the stock market change so much over time? And the authors of this paper suggest that part of the reason can be these, these consistent political crises through time. We've seen the stock markets do tend to react negatively to political crises and in particular to wars.

Bond markets, and this is a tricky piece, because bonds are typically the safe assets in a portfolio. Bond markets have done even worse, which flips that conventional wisdom on its head a little bit, at least in the historical data. Over the last 121 years, five countries have experienced negative real bond returns for the full period where no country has had negative real stock returns for the full period. And those negative real long term bond returns are largely associated with wars and their aftermath, in the first half, the 1900s.

German bonds lost all of their value during the hyperinflationary period of 1922 to '23. Even the US bond market has had major periods of long term under performance in and around war time. Starting in December, 1940 us long-term government bonds lost 67% of their real value and did not recover in real terms until 1991, pretty scary. Similarly, UK long term government bonds started dropping in October, 1946, lost 74% of their real value by December, 1974 but really just a slow, steady decline. Because that's a cumulative number. And again, they didn't recover fully until December, 1993.

Now my point there is not a cautionary tale about don't own bonds. I think it's just a reminder that diversifying across stocks and bonds and maybe diversifying bonds globally makes a lot of sense. US and UK stocks had positive, real returns over the same period where those bond markets had big negative returns. Stocks and bonds have historically been imperfectly correlated. And even though bonds have had big drawdowns over some time periods, they're still generally less volatile than stocks. It's a reminder, I guess, even though bonds are generally less volatile than stocks, they are not necessarily a safe asset in real terms, over long periods of time.And in historical data, a lot of those big drawdowns and big negative real returns over long periods of time have started with worse.

The idea of panicking, getting out of stocks, and going to bonds, for example, maybe it doesn't sound so good, at least not in the historical data. Although, a 60/40 portfolio performed quite similarly to stocks, I guess, because of the rebalancing over those historical periods where bonds did quite poorly. So, it's not like the 60/40 portfolio didn't work just because bonds had negative real returns. It still had less volatile returns than a 100% stock portfolio, just not as good as returns would've been if bonds had been like they have been for the last 30 or so years.

Wars can be devastating, particularly for the individual countries on the losing end. This might be the most important point of this whole segment. I think investors have to understand that the worst historical global stock market returns have occurred in peace time, and those peace time crashes have occurred more frequently than major wars than market declines associated with major wars. We mentioned global stocks lost 31% in real terms in World War I, 12% or 15% ... I think it's 15% in World War II. But they lost 54% in the Wall Street crash of 1929, 47% in the 1973 oil shock and recession, 44% in the 2000 internet bust, 41% in the global financial crisis in 2008. So, you look at that data, and it's like, if you're panicking about market volatility because of a war, maybe you shouldn't have been in stocks in the first place. War is not what causes the worst cases of market volatility and drawdown.

Cameron Passmore: Exactly.

Ben Felix: But if a war is your tipping point, well, you're probably in too aggressive of a portfolio in the first place. The worst declines that you can expect are not associated with war.

Cameron Passmore: Exactly.

Ben Felix: So, maybe another one of those reasons to revisit asset allocation. But in terms of worrying about market declines, wars are, at least historically, not the biggest risk factor. It's been financial crises and recessions that has driven most of that.

The last big lesson to remember is that while stock returns have been volatile throughout history, they have been reliably positive in the long run. And for a globally diversified investor, the ride has been relatively stable compared to investors concentrated in individual countries. Global stocks have returned a real inflation adjusted of 5.2%. In historical data going back 121 years, inflation over that period has been about 3% in Canada and the US. I'll call it an 8% nominal return, 8% per year on average, which is huge.

That's higher than what we expect in our current financial planning expected return assumptions. And that return is despite two large scale global wars, the cold war, multiple civil wars around the world, revolutions, economic depressions, and pandemics, including the current one. In total, there are 487 political crises in that database that I mentioned earlier from 1918 to 2017, that the database has been updated since that paper was written. So 487 crises, not to mention all the other crazy stuff that goes on in the world. Here we are, and market returns have looked just fine.

A common response that I hear to long term data like that is, "Well, I don't have 121 years. I need my portfolio to maintain its value now." Well, if you need a portfolio to maintain its purchasing power at all times, then you should be in cash or GICs. But I think it's important to remember the diversification between stocks, and bonds, and across countries, and exposure to multiple sources of expected return, like the small gap in value that I mentioned earlier has dramatically dampened volatility, reduced the magnitude of the most extreme drawdowns, and increased recovery time after crises. It's more of a story about diversification than it is about not being in the market.

And that's what Larry Swedroe gave us such a good answer to that question, when he was a guest on our podcast a while ago, the first time that he was a guest. He's been on twice, but we asked him that question, that sometimes people tell us that they don't have time to wait for factor premiums. And Larry says, "No, that's absolutely backwards. You need more sources of expected return. They're short of your time horizon is, not the other way around."

We're faced with risk in investing. People talk about risk a lot. Risk can be quantified with known probabilities. And that's probably not even the best description of risk in financial markets, because there's a lot of uncertainty as well. But maybe generally investing in a global index tilts us a little bit more toward risk than uncertainty, especially if you own the whole market. Uncertainty cannot be quantified. I think political risk and war are more uncertainty than they are risk. You can't really plan for that kind of thing with asset allocation.

In Neil Ferguson's paper, the title of the paper is Earning from History? Financial Markets and the Approach of World Wars. It's the paper that I mentioned a couple times, Neil Ferguson finds, similar to the quote I mentioned earlier, that changes in military technology and government regulation ensure that one could never be certain that the next war would have the same financial impact as the previous war.So again, it's kind of like, can you use the onset of war to time the market? No, because they're always different, but that doesn't mean that there are no lessons to be taken from history. I thought this was very interesting. This is Neil Ferguson's commentary. His big three lessons that he gives at the end of his paper are that major wars can arise even when economic globalization is far advanced. The longer the world goes without a major war, the harder one becomes to imagine, and I think people can probably relate to that now. When a crisis strikes complacent investors, it causes much more disruption than when it strikes battle scarred ones. That's his observation looking at the behavior of investors from World War I, World War II, and the Korean War, and maybe the Cold War too. He looked at all those different events. This is a quote from Keynes, but I found it in Neil's paper.

In 1937, Keynes wrote about respond to uncertainty, the non-quantifiable type of risk. And this is in 1937, so the years before the Second World War. Keynes said, "We largely ignore the prospect of future changes about the actual character of which we know nothing." You can't predict the future. " We assumed that the existing state of opinion as expressed in prices and the character of existing output is based on a correct summing of future prospects. Knowing that our own individual judgment is worthless, we endeavor to fall back on the judgment of the rest of the world which is perhaps better informed. That is we endeavor to conform to the behavior of the majority or the average."

I don't know, that sounds like index funds to me. When you take what Neil said and what Keynes wrote, it's kind of like you take only as much risk as you can handle and as much risk as you can handle even in bad times, or especially, in bad times. But then the other piece of that is don't forget the bad times will come. You do your asset allocation, you do your financial plan, and we do this using Monte Carlo simulation and we do it using conservative expected return assumptions. But you can't forget the bad times will come, all you can do is plan for them. When the bad times do come, you know that you plan for them, and therefore, you can keep your head down and diversify. And you can't time the market so you keep your head down and continue doing what your plan dictated that you should be doing to meet the objective.

Cameron Passmore: That was a big revelation that Dave Goetsch had that he talked about with us when he was on at the beginning of the pandemic with Dr. Summers.

Ben Felix: Yeah, that's right.

Cameron Passmore: Once he realized and learned to embrace volatility, and like you said, hug volatility, and once he had that mindset shift, he was able to withstand these inevitable episodes.

Ben Felix: I hope our commentaries helped a little bit because I know some people are nervous because we've gotten some phone calls and emails and I've had some conversations. So I don't think it's not us feeling nervous and wanting to do this. I think we're responding to how other people might be feeling. I also wanted to give a couple notes on what Dimensional is doing. They gave us a briefing on how they're thinking about and handling the situation, and I thought they had some really insightful points. Dimensional, the theme of their talk was... they didn't say this, but it's what I took away from it. Risk management happens before a crisis. Well, it's what I just said with asset allocation and with planning. Risk management happens before a crisis, not after.

Cameron Passmore: Not during, exactly.

Ben Felix: When you ask Dimensional, "What are you doing to respond to this?" The answer effectively, and I'll say what they actually said, but the answer is effectively, "Oh, well, we already did it." The risk management happened before the risk showed up, which is exactly how it should be. They decide whether to enter a market, and particularly, an emerging market based on things like the costs and frictions associated with accessing the market, the liquidity of the market, the level of regulation in the market at the individual exchanges in the country, the listing requirements of the exchanges, the accounting standards in the country, the property rights, and the treatment of foreign investors. Those are some of the big things that are likely more, but those are the big things that Dimensional looks at to decide should we enter this country?

Now, Russia, according to Dimensional, has relatively lax listing standards. So Dimensional said, "Okay, we're not going to go direct onto the St. Petersburg or whatever the stock exchange is and buy securities. We're going to go through ADRs." Those are American Depository Receipts, and those are listed. The ones that Dimensional owns are listed on the London and U.S. stock exchanges. That means that the ADRs have to meet U.S. and UK listing standards, not the emerging market countries that they're originated from, or that the ADRs are representative of, so that makes that go away. Dimensional doesn't need to buy Rubles to manage their positions in Russian stocks because they're transacting on U.S. and UK exchanges. That makes some of that go away.

Now, the other thing that I thought was really interesting about their talk, and I didn't know this prior to hearing it, is that their weight in Russian securities is small relative to Cap weights. In the benchmark like ETF portfolios, and I looked at Vanguards and iShares' emerging markets portfolios, they're at 2.5 to 3% in Russia. Dimensional, in 2014, strategically decreased, not based on a prediction, not we think Russia's going to underperform, but they decrease the weight due to concerns about property rights. And they did that underweight in 2013. It's been 1 to 1.5% in Russian securities, in Dimensional's emerging market strategies since since 2014.

But, like I said, I checked Vanguard and iShares. They are indeed still at 2.5 and 3%. I can't remember which one was 2.5 and which one was 3, so Dimensional is underweight relative to Cap weights. Again, that's back to my comment of when do you do risk management before crisis happens? They did that and they also said that they are currently evaluating sanctions and that may result in further changing their exposure to Russia. But their other big comment was, "This situation is not good. We're monitoring it closely, but at the end of the day, it is a relatively small allocation in the emerging markets portfolio, which is a relatively small allocation in the overall global portfolio." But I didn't look at the percentage of the total portfolio, but 1.5% of the emerging markets portfolio is going to be well under a percent in the global portfolio. Anyway, I thought that was just insightful, but it's back to the general concept of managing risk before a crisis.

Cameron Passmore: For me, the two key takeaways, one is that manage risk ahead of time and markets have volatile periods outside of war time.

Ben Felix: The biggest drops in global markets have been outside of war time, they've been in peace time. War does increase volatility, it is associated with negative returns on average, but not always and not in all countries. So you can't really use it to time the market. But then the other big piece of this is that the global market has been remarkably resilient despite all the crazy stuff that we've seen throughout history or, well, for the last 121 years anyway.

Cameron Passmore: Good information.

Ben Felix: All right, yeah, so that's it. I hope it's helpful for people. It was definitely helpful for me to go dig through all of that data. I learned a few things like the point about the biggest market crashes being in peace time as opposed to during wars. I thought that was an interesting takeaway. Likewise, with St. Petersburg stock exchange outperforming the U.S. from 1865 to 1917. Another new one for me. Anyway, we hope this has been helpful.

Cameron Passmore: Exactly. Thanks.