Thematic, innovative and bespoke investment solutions and thinking
A review of Megatrend 

Mean Reversion of Wealth-to-GDP - February 2024 Update

I'm always thinking about losing money as opposed to making money. Don't focus on making money, focus on protecting what you have.

Paul Tudor Jones


Strictly speaking, Mean reversion of wealth-to-GDP is not a megatrend. Rather, it is what I would call the expected aggregate outcome of the megatrends we have identified so far. In this paper, I shall spare you from any details as to why wealth-to-GDP must mean revert in the years to come. I have written extensively why that is and don’t intend to revisit the topic today.

Having said that, if wealth-to-GDP is well above its sustainable mean value, the implication is that wealth (when measured as a percentage of GDP) must decline. Hence, the combined effect of ‘our’ seven megatrends will, if the underlying economic theory is correct, be negative, even if a few of the megatrends we have identified may have a positive impact on wealth.

I should also stress that the long-term mean value changes over time. Last time I came across some academic work on the subject was in 2011, and the mean value, when measured over the entire post-industrial revolution period, was about 380 in the US and somewhat higher elsewhere. Now, with another 12-13 years under the belt, one would expect the mean value to have increased, as wealth-to-GDP has been significantly above 380 in recent years i.e. wealth-to-GDP will need to revert by a smaller amount, provided wealth-to-GDP does not begin to increase again.

Finally, before I start, allow me to remind you of an important detail. When I write about wealth-to-GDP, I tend to focus on US wealth-to-GDP, and there are at least a couple of reasons for that. First and foremost, US wealth-to-GDP is further away from its mean value than it is elsewhere, i.e. the wealth impact is likely to be more dramatic there. Secondly, US financial markets are by far the most important financial markets worldwide. A significant correction in US wealth is therefore likely to have a meaningful impact on financial markets in the rest of the world.

Continue reading

We publish investment strategies and opportunities in our research papers. This research paper is available to professional investors as part of ARP+ subscription.
Find out more
Already a subscriber? Login

The latest numbers

Our inaugural paper on Mean Reversion of Wealth-to-GDP was published in February 2021 and started as follows:

“Wealth in society when measured relative to GDP is, for reasons I will explain in a moment, long-term stable. The challenge facing investors now is that wealth has risen much faster than GDP over the past 25 years and is now well above the long-term mean value in most countries. Wealth-to-GDP reverting to the mean is thus long overdue.”

Three years later, US wealth-to-GDP continues to be way ahead of itself, although the ratio has moderated somewhat more recently. The combined effect of a stagnant numerator (flattish wealth since late 2021) and a growing denominator (robust GDP growth in recent years) has, to some degree, deflated US wealth-to-GDP. It peaked at 615 in 3Q21 and now stands at 547 (Exhibit 1).

Exhibit 1: US wealth-to-GDP ratio (%)
Source: Federal Reserve Bank of St. Louis

Having said that, there is still a long way to 380 – the long-term mean value of US wealth-to-GDP if you include data going back to the beginning of the industrial revolution. However, the Fed doesn’t provide data prior to World War II, and older data is often criticized for being less accurate. Although there is some truth to that, the data is accurate enough to support the theory which states that:

(i) sooner or later, wealth-to-GDP always reverts to its mean value (assuming an open market economy);

(ii) the mean value is independent of the level of economic development, i.e. the argument that the inflated level of wealth-to-GDP can be justified because of rising living standards doesn’t fly; and

(iii) the mean value is not the same in every country but is a function of how effectively capital can be deployed in the country in question. The US is the most capital-efficient country – thus, the US mean value is lower than elsewhere.

More specifics

Credit Suisse used to provide the most detailed information on wealth worldwide; however, with the bank having been absorbed by UBS, it is now up to them to pick up the baton. Last year was a first for UBS with the publication of Global Wealth Report 2023, and I would expect the 2024 report to be published relatively soon.

Last year’s paper was a bit unusual in the sense that wealth had fallen significantly the year before (2022) – only the second time it has done so in the new millennium (Exhibit 2). As you can see, it was financial wealth that took a beating, reflecting the difficult conditions in most financial markets in 2022. Non-financial wealth actually increased by a couple of percent that year. Now, with the 2024 report around the corner, I would expect both financial and non-financial wealth to have risen in 2023.

Exhibit 2: Annual change in real global household wealth and its components, %
Source: UBS Global Wealth Report 2023

Given the strong performance of equity markets last year, the 2023 increase in household wealth was actually surprisingly modest – at least in the US. This is obvious if you take another look at Exhibit 1. I should also point to a significant difference between the numbers in Exhibits 1 and 2. Where the Federal Reserve Bank’s numbers in Exhibit 1 are nominal, UBS’s numbers in Exhibit 2 are inflation-adjusted. For many years, this did not make much of a difference, but it certainly does now.

If one digs a bit deeper and look at the change in wealth by country, an interesting picture emerges (Exhibit 3). As the authors of the UBS paper point out, the countries that lost the most wealth in 2022 were often those that made the biggest gains the year before. This is another nail in the coffin for all those who do not subscribe to the theory behind it all. Mark my words – wealth-to-GDP will continue to fall – not necessarily every year, but the trend line will be down until we have fully mean-reverted. Wealth creation is a byproduct of GDP growth, and it is absurd to think one can outgrow the other forever. The problem I have as a forecaster is that the underlying theory provides zero guidance on timing. Therefore, I can end up being wrong for an awfully long time before I am finally vindicated. And that is also the reason you should never allow it to control your portfolio construction. Keep an eye on it, but do not allow it to take charge.

Exhibit 3: Change in real wealth per adult (in USD) by country, 2022 vs. 2021
Note: Biggest gains and losses only
Source: UBS Global Wealth Report 2023

The various components of wealth

In most countries – certainly in all developed countries – the most important components of wealth are (in no particular order):

- pension savings;

- investments (ex. pension savings) in various financial instruments;

- private market investments (mostly family-owned businesses); and

- real estate.

The mix between the four varies a great deal from country to country. Take for example Germany vis-à-vis the UK. In Germany, it is far more common to live in rented accommodation than it is in the UK. Therefore, real estate accounts for a higher proportion of household wealth in the UK than it does in Germany.

A similar picture emerges when one looks at the mix of financial assets. Where Germans typically prefer to keep their savings in savings accounts and invest in bonds rather than equities, the opposite is the case in the UK.

Market cap-to-GDP by country is a reasonably good proxy for how popular it is to invest in equities in various countries. As we all know, US households are the ultimate equity investors and have been in love with equities for as long as I can remember. A negative tone in equity markets may therefore have a bigger impact on American consumer spending than on consumer spending elsewhere.

In fact, equity investing is so important to many US households that the government changed the mandate of the Federal Reserve Bank some years ago. It used to be solely about price stability but is now about maximum employment as well – another word for robust economic growth. Effectively, the Fed has been mandated to stimulate the US economy, even if inflation is not precisely where they want it to be. Therefore, it shouldn’t come as a major surprise that market cap-to-GDP is higher in the US than in any other country (Exhibit 4).

Exhibit 4: Market cap-to-GDP by country
Sources: Crescat Capital, Bloomberg

Going back to the four wealth components listed above, an interesting ‘amalgamation’ between pensions and private market investments has unfolded more recently. Allow me to explain. Frustrated by relatively poor returns in listed equities (away from the Magnificent 7), many pension funds have allocated significant amounts of capital to private equity (PE) in recent years; in fact, so much capital that it has become a real problem for PE managers to deploy that capital in investments which can generate the sorts of returns that (i) investors expect and (ii) justify the illiquidity of PE.

Nobody knows how this is all going to end, but if PE funds struggle to deliver the returns investors expect, not only will it affect the value of your pension savings, but it could also affect the valuation of many family-owned businesses which, to a significant degree, are valued on the basis of the multiples that PE transactions are conducted at. Therefore, falling multiples in PE transactions could start a rather unpleasant chain reaction.

Investment implications

As I said earlier, keep an eye on this dynamic, but do not allow it to take complete control of your portfolio construction, which is precisely how we manage it at Absolute Return Partners. In our new fund – the ARP Megatrend Fund – we deliberately keep the equity beta well below one, but we still take plenty of equity risk.

There are (at least) two simple ways to outperform the broader equity markets. You can either construct your equity portfolio so that the equity beta is above one, or you can pick corners of the market with an equity beta below one that you, for idiosyncratic reasons, expect to do well – both in benign and in more challenging conditions. The difference between the two methodologies is that the former method requires for the equity market to go up for you to make money, whereas the latter can make you money even when conditions are challenging.

In the ARP Megatrend Fund, we have made a number of investments that keep the equity beta in the portfolio well below one. As I write these lines, the fund’s equity beta is 0.73. In other words, if the MSCI World Index (the fund’s benchmark) were to fall 10% tomorrow, we would ‘only’ expect the fund to fall by 7-8%. In general, we are very happy with the fund’s up/down capture characteristics so far. We have participated less in down markets than we have in up markets, which is a very attractive feature.

You may want to put your money into a fund with an even lower equity beta, which is effectively what market-neutral equity long/short funds offer; however, the price you typically have to pay when investing in those types of funds is limited upside participation. Equity long/short funds typically find it hard to keep up with the indices when markets rise (which they do most of the time). Adding to that, many of them also charge outrageous fees, which I find hard to stomach.

One last observation: In the ARP Megatrend Fund, we have 11 so-called baskets – investment themes resulting from the seven megatrends. At present, the baskets enjoying the lowest equity betas are Japanese Banks (-0.10), Uranium (0.30) and Carbon (0.33). On the other hand, the baskets with the highest equity betas in the fund are AI (1.58), Lithium (1.54) and Renewables (1.44). Although it is not our intention to minimise the equity beta (as that would severely impact the upside), we are constantly on the lookout for attractive investment opportunities that can bring the consolidated equity beta a bit further down.

Niels C. Jensen

14 February 2024