Paul: Welcome to the fourth episode of Index Ideas from FTSE Russell. I'm Paul Amery, your podcast host. In the podcast, we explore index ideas that can help address real-world investment themes and challenges. The concepts that we discuss in the podcast are not investment advice. They represent transparent, systematic approaches that may be developed into indices and made available for broad public use. Any reference to potential strategies is therefore intended for informational and educational purposes only. In this episode, I'm joined by Alberto Allegrucci, who is a quantitative financial researcher at FTSE Russell. Alberto is here to talk about some research he and his team have recently published on one of the most important topics for investors, asset allocation. Alberto, welcome to the podcast.
Alberto: Thank you very much, Paul.
Paul: What is the 60/40 asset allocation model and why does it need a rethink?
Alberto: So the 60/40 model, it is a traditional portfolio mix whereby an investor invests money 60% in equities and 40% in bonds. And it's a really famous asset allocation model—perhaps the most famous one—because equities, you know, in general, they drive the growth of the portfolio, and bonds provide some kind of income and diversification, especially during bad times. And now this portfolio, this kind of asset allocation in the last 20 years performed really well. But now we need to think about it a little bit more, because the reason why this asset allocation performed so well was because of the particular economic environment in the last 20 years: there was a low-rate environment and low inflation. Right now, that’s not the case. There’s a little more volatility and bonds have started to become more correlated with equities. They're not hedging their risk as effectively as before.
Paul: So why was low inflation critical to the success of 60/40 in the past?
Alberto: When inflation is low and stable, that creates a favourable environment. There is research that shows that in this kind of economic environment, the correlation between bonds and stocks tends to be negative, which is good for asset allocation, because, when one asset does badly, the other one makes up for it. So, it creates an environment in which this kind of investment is pretty stable and performs pretty well. For bonds, for example, when interest rates are low and predictable, prices are well-supported. And, for equities, during this kind of environment the policy has been quite expansionary, with the low [interest] rate like it has been in the last 20 years. That supports a little bit the valuation of the equities themselves.
Paul: So after 2022, we were coming out of the Covid period and inflation took off. What happened to the bond-equity correlation from that point onwards?
Alberto: Yes, after 2022, there was a bit of a shift, inflation went up. And central banks, they had to tighten quite quickly their monetary policy. Bonds and equity, they sold off at the same time. And they kind of break the traditional negative correlation that has been there for the last 20 years. And both asset classes--bonds and equities--delivered one of the worst returns in decades.
Paul: So we switched from a negative to a positive correlation between the two.
Alberto: Exactly. That's what is happening right now.
Paul: So if we go further back in history what's the typical range? You know, can we say that there's a typical range for bond-equity correlations?
Alberto: Yeah. In general, we went back in history until like 1930. There are different macro environments, but generally when the inflation goes up, we have a positive correlation and it goes up until like maybe 0.5 and recently it has been negative and it was -0.2, -0.3, which is what made the 60/40 effective.
Paul: So as a result of this observation on correlations, you and your team have worked on a what you call a balanced macro asset allocation framework. Could you explain what you've defined as the growth versus inflation quadrant framework. So, what are the four regimes?
Alberto: So the growth versus inflation quadrant framework is a well-known framework. And it maps the macro environment into four quadrants, depending on whether growth and inflation are rising or falling. And the idea is that you have different asset classes that perform differently depending on the quadrant that we're in. And the four quadrants are: rising growth, falling inflation, which is good for risky assets like stocks; rising growth and rising inflation—here commodities and real assets tend to perform a little bit better; falling growth, rising inflation—this is tough for both stocks and bonds, but we found that precious metals tend to perform a little bit better; finally, falling growth and falling inflation, when bonds typically do well, whereas equities might suffer in this kind of environment.
Paul: So in each quadrant there are a set of assets that would be preferred.
Alberto: Yeah, exactly. So, in each quadrant we map all the assets that we have available that we can track at FTSE to each quadrant. And yeah, we started from there to build our asset allocation and balanced macro.
Paul: So do we need to know which quadrant we're in at any particular time?
Alberto: So in general, knowing what quadrant we are in is good because it helps the asset allocation, the positioning of the portfolio. But this is really hard to do. It's really hard to predict what quadrant you're in. And if you make a mistake that can be costly for your portfolio allocation. So, in our case we took a different approach. There are other types of allocation that don't require you to know where you are in. And you just need to know like what kind of assets perform well in each quadrant. And then you invest a little bit in all of them, trying to maintain the allocation balanced so that you're covered. You know, you try to be covered in every kind of scenario.
Paul: So could you go into a bit more detail and explain what the balanced macro framework is? How do we decide how much to allocate to each set of assets, and do we vary that over time and if so, how?
Alberto: Yeah. So, when we build a framework, we want it to be view-agnostic because it's difficult to predict regimes. So, we try to invest a little bit in every kind of asset class for each quadrant. And the way that we do it is a two-step process, trying to balance both the macro risks and also the risks baked in in each asset class.
So firstly, for the asset classes grouped in the same quadrant, we use equal risk contribution and we estimate risk using both volatilities and correlations between asset classes. And once we did that for each quadrant we ended up with four portfolios. And we need to mix them up. And we aggregate those four strategies using again equal risk contribution to balance the risks. But we don't use correlation because assets across quadrants tend to have a shift in correlation which is difficult to predict: for example, stock and bonds. It changed quite quickly. And we don't want to use correlations that are difficult to predict. So we are using just volatilities to estimate the weights that we want to have for this asset allocation.
Paul: So could you give an example of you know you talked earlier in the podcast about the shift in regime we saw in 2022. What has the framework been suggesting for that particular quadrant we're probably in at the moment. And what asset classes does it come up with?
Alberto: So that's kind of the beauty of it. As a result, the way that we allocate it didn't change too much. But because, we showed that when inflation is rising, commodities and precious metals, they do quite well. The portfolio would reinvest in those asset classes. And that would smooth out a little bit the downturn resulting from equities and bonds.
Paul: So why did you and your team come up with this particular approach? Was it demand from clients, was it internal discussions? Where did this idea come from?
Alberto: We started a little bit from an internal discussion. We thought that after 2022, we wanted to have something a little bit more comprehensive rather than offering you know, like 60/40. At FTSE Russell, we have a broad set of indices. And these kinds of strategies would be rather easy to implement. And also, because we have so many indices, we can offer some customisation too. You know, at the start we didn't want to be in the business of predicting the regime we are in.
Paul: So it's a flexible framework that doesn't require any prediction. You know, we're obviously not in the business of giving investment forecasts.
Alberto: Exactly. The only thing that we do is estimating the covariance matrix, which is kind of common. But we also have a version of the balanced macro in which we just equal weight within quadrants and across quadrants to make it a little bit simpler. It's less balanced, but it's simpler, and it doesn't require even the estimation of the covariance matrix within between asset classes.
Paul: Okay. You mentioned earlier that you've looked at bond-equity correlations as far as in the 1930s. In your back tests of the balance macro model, what did you find with the risk and return outcomes?
Alberto: So we ran a back-test starting from the 1970s and up until 2024. And in our balanced macro, the actual equal risk contribution version, we found out that the performance was a little bit less than investing only in equities (mainly in American/US stocks). But we wanted to achieve a better risk-adjusted return. And yes, I think we have like 7.3% per annum (return) against like 11% for investing only in equities. But then when you look at volatilities, the volatilities are much lower even proportionally. So, when we go and see the risk-adjusted return, in that the performance is much better. Just to give you some numbers, the balanced macro’s risk-adjusted return during the full back-test was 1.35, and instead investing only in US large caps would yield a risk-adjusted return of 0.74. If we look at the maximum drawdown, the balanced macro, over the full history of the back-test, had a maximum drawdown of -16.7% against around -50% for the US large cap [equities]. So, from a risk perspective I believe it's a much better option.
Paul: And for listeners who would like to find out more about the model, how it works, about your research into this area, is there somewhere they could go on the FTSE Russell website to read more about it?
Alberto: Yeah, there is a published blog and a paper and they are on our website. And obviously if there are more questions, we are always available to discuss, we will be happy to discuss.
Paul: And how can clients in general implement this model? Is it an index? Is it just an index idea? You know, where are we with the production?
Alberto: Right now we have a back-test and it's an index of indices. So, all the asset classes that are part of this framework, they are backed by indices that FTSE Russell publishes. And then when the client wants to implement it, whether [the client] does it by buying an ETF for more wealth manager type of clients or, you know, maybe in a future, like levering up, that's, you know, something that we are always happy to discuss—how to best accommodate these things better.
Paul: Alberto, thank you very much for taking the time to speak to me.
Alberto: Thank you Paul.
Paul: That's it for this episode. If you've enjoyed the conversation, then please do follow us and give us a rating or review on your podcast app of choice. And if you'd like to get in touch with the show, you can do that via the email
[email protected]. But for now, from me, Paul Amery, goodbye.
