This article summarises Leithner & Co.’s assessment of three key commonly-cited macro-economic risks: (1) increasing inflation, (2) stagnant or declining economic growth and (3) rising interest rates. By analysing a long series of data compiled mostly by Robert Shiller, I show that:
- In isolation, waning growth (as measured by Gross Domestic Product) as well as waxing inflation (Consumer Price Index) and long-term rates of interest pose little threat to investors: they cause the S&P 500’s gains to moderate, but on average its returns remain positive.
- Under certain conditions, the confluence of these macro-economic variables crushes investors’ returns. In particular, sharply rising rates during “stagflation” (that is, periods of sluggish growth and rising CPI) typically generate significant losses.
Stagflation has prevailed in the U.S. since June 2020. Investors should therefore beware: if past is prologue and Australia cannot indefinitely resist economic and financial infections from abroad, for how long can recent stellar returns last?
A Short Note on a Long Series of Data
This article relies primarily upon monthly series measuring key variables in the U.S. since January 1871 compiled by Robert Shiller for his book Irrational Exuberance (1st ed., Princeton University Press, 2000) and updated thereafter. Details of these data appear in that book and on his website. For particulars about the GDP data I’ve merged with Shiller’s data, see Carmen Reinhart and Kenneth Rogoff, This Time Is Different: Eight Centuries of Financial Folly (1st ed., Princeton University Press, 2011) and Reinhart’s website.
To what extent do inferences from the U.S. over the past 150 years apply to Australia today? We can’t know for sure because corresponding data of comparable quality have existed in this country only over the past few decades. Yet Australian financial markets have long reacted to economic developments in the U.S.; similarly, the Australian economy has long imported macro-economic developments from overseas. I therefore assume that to a significant extent conclusions from these American data extend to Australia – but, as in all things economic and financial, caution and scepticism are warranted.
GDP and Returns
Why do so many investors obsess about the economy and its prospects? They assume that its growth (conventionally measured by changes over time of GDP) reverberates to shareholders’ returns.
This expectation is simple and plausible. Yet reality isn’t nearly so clear-cut. Several major analyses have examined the effect of GDP’s rate of change upon stock markets’ returns in a range of countries. Results have been anything but straightforward: they depend upon the definition of “short-term,” “long-term” and the like, as well as the period of time and country analysed; accordingly, even modest tweaks of an analysis have significantly changed its results. Figures 1a-1b provide an example.
Figure 1a summarises the relationship between GDP’s rate of growth and the S&P 500’s total (that is, including dividends) rate of return. (This and subsequent graphs describe results during the ca. 1,800 12-month periods since 1872. Similar results also occur over five-year periods since 1876, but for the sake of brevity and simpler graphs I’ve omitted them.) I ranked the data by GDP’s growth and divided them into four equal (by number of observations) portions. During 25% of these months (Quartile #1), growth varied between -23.1% and 3.0% (during the depths of the Great Depression it collapsed more than one-fifth); 25% of the time its rate of growth varied between 3.0% and 5.3% (Quartile #2), and so on. In 1940-1944, as America’s wartime production skyrocketed, annualised rates of growth vaulted above 28%.
Figure 1a: Total Return, S&P 500, by GDP, January 1872-June 2021
Although many investors obsess about the economy, it’s easy to see why some ignore GDP and predictions about its future course: regardless of its rate of change, on average the S&P 500 generates positive returns. Since January 1872, the Index’s 12-month rate of total return has averaged 8.6%. As growth accelerates, the Index’s return tends to lift – except in Quartile #4, where it falls. As a result, overall the relationship is curvilinear.
This is mostly because periods of strong GDP growth have also tended to be periods of high consumer price inflation – and as we’ll see shortly, sharp rises of CPI clobber the S&P 500’s return. In short, economic growth tends to boost returns – unless consumer price inflation runs rampant.
Given the huge variability over time of CPI (see the next section), for each month since January 1872 I also calculated GDP’s real (that is, net of CPI) 12-month percentage change, as well as the S&P 500’s real total 12-month rate of return. Figure 1b plots the results: an extremely strong, positive and curvilinear relationship.
Figure 1b:Real Total Return, S&P 500, by Real GDP, January 1872-June 2021
These results should trouble bulls. During the past five years, GDP’s rate of growth has averaged 3.0% per year; CPI-adjusted, however, it’s averaged just 1.0% – and since June 2020 it’s averaged -2.5%. In Figures 1a-1b, these observations cluster towards the upper bound of Quartile #1 and the lower bound of Quartile #2. It’s true that in nominal (unadjusted for CPI) terms GDP rose 6.5% in the 12 months to September; but that’s by far the strongest rate of growth since the GFC – but in real terms it’s closer to 0%.
Yet the S&P 500’s recent returns have greatly exceeded Quartile #1-2’s means. If past is prologue, investors should anticipate much lower returns.
CPI and Returns
The Consumer Price Index, says Investopedia,
is probably the most important and widely watched economic indicator, and it’s the best-known measure for determining cost of living changes which, as history shows us, can be detrimental if they are large and rapid … By knowing the state of consumer prices, investors can make appropriate investment decisions.
Figure 2 summarises the relationship between the CPI’s 12-month rate of change and the S&P 500’s 12-month real rate of total return. During 25% of the months since the early-1870s (Quartile #1), CPI has shrunk (and during the depths of the Great Depression it collapsed almost one-fifth); 25% of the time its rate of growth has varied between 0% and 2.3% (Quartile #2), and so on.
Figure 2: Real Total Return, S&P 500, by CPI, January 1872-June 2021
In 1941-1942, as production soared and Depression-era unemployment disappeared, consumer prices vaulted to annualised rates well above 10%; and in 1946-1947, as the economy transitioned to peacetime, CPI’s 12-month rates of growth neared 20%. Consumer price inflation also zoomed above 9% during the Korean War, rose significantly (above 5%) during the late-1960s and remained elevated (and as high as 14.3% per year) until the mid-1980s. Indeed, only in the early-1990s did it fall sustainably below 3%. Its recent rate of increase (above 5% since June) is, apart from a brief blip in 2008, the highest in 30 years.
It’s also easy to see why many investors should ignore CPI and predictions about its future course: at rates of change below ca. 4.5 per year, which occurs 75% of the time, the S&P 500 generates excellent returns. But if CPI rises at a rate of more than ca. 2.3% per annum, average returns fall; and if it rises above 4.5%, they virtually disappear.
This, too, should also concern bulls. Given recent months’ “prints” (in September the 12-month rate of increase was more than 6%), there are plausible grounds to suppose that CPI’s rise can continue to exceed 2.5% per year. But it’s hardly certain that it will remain above 4.5% per annum. Should it do so, investors can expect the S&P 500’s returns to stagnate.
Long-Term Rates and Returns
“The investment community and financial media tend to obsess over interest rates and for a good reason,” says Investopedia. “Understanding the relationship between interest rates and the stock market can help investors understand how changes may impact their investments …”
Figure 3 summarises the relationship between long-term (10-year) rates’ 12-month percentage rate of change and the S&P 500’s 12-month real rate of total return. (The results for CPI-adjusted and unadjusted returns are almost identical.) Importantly, it assesses the effect upon returns of long-term rates’ percentage change – not their level – during each 12-month period.
Figure 3: Total Return, S&P 500, by Percentage Change of Long-Term Rate, Jan 1872-Jun 2021
As an example, in April 2019 the ten-year yield was 2.53%; by April 2020, it had collapsed to 0.64%; therefore its 12-month change in the latter month was (0.64% – 2.53%) ÷ 2.53% = -73.9%. That’s its biggest fall on record. In April 2021, in contrast, the rate was 1.64%; hence its 12-month percentage rate of change was (1.64% – 0.66%) ÷ 0.66% = 148.5%. That’s its biggest increase on record.
During one-quarter of the 12-month periods since 1872 (Quartile #1), rates have fallen – sometimes sharply; another 25% of the time (Quartile #2), they’ve decreased slightly; during one-quarter of these intervals they’ve been mostly stable (Quartile #3) and the rest of the time they’ve risen – sometimes rapidly (Quartile#4).
Like rates of change of CPI and GDP, so too long-term rates’ short-term fluctuations: considered in isolation, investors would do well mostly to ignore them. Regardless of their direction and magnitude of change, on average the S&P 500 generates positive returns. Nonetheless, the more strongly in percentage terms that rates rise, the lower is the return.
What about Stagflation?
Thus far I’ve analysed all 12-month periods since January 1872; I’ve also considered CPI, GDP and long-term rates mostly in isolation. In reality, of course, these variables influence and counteract each another. For example, strong boosts of GDP have often accelerated CPI’s rise; sharp increases of CPI, in turn, have tended to cause long-term rates to lift – and GDP’s growth to decelerate or reverse.
Of course, exceptions to these generalisations abound. Most notably, “stagflation” denotes the confluence of low GDP growth (stagnation) and high CPI growth (inflation). According to Google Trends (25 November), in October searches for this term neared (88%) the all-time highs they scaled during the GFC (Figure 4).
Figure 4: Google Trends for “Stagflation” since January 2004
For two reasons, this current interest is understandable: first, since mid-2020 stagflation as I define it has prevailed; second, historically, during such periods investors have often sustained significant losses. I selected the ca. 450 observations in Quartile #1 in Figure 1b (which I dub “stagflationary” periods), ranked them by the 12-month percentage rate of the long-term rate of interest and divided them into four equal (in terms of numbers of observations) groups. For each quartile I computed the average of the S&P 500’s real rate of total return. Figure 5 summarises the results.
Figure 5: Real Total Return, S&P 500, by Long-Term Rate during Periods of Stagflation
Stagflation hobbles the S&P’s 12-month total real return. In this portion of the dataset its average is 2.8%; in the full dataset it’s 8.6%. During periods of stagflation when the long-term rate falls sharply (Quartile #1), the Index’s average real return exceeds 17% per year. But when rates remain stable (Quartile #2) it slumps to an average of 3.3%.
The biggest problem is that rates rise during half of these stagflationary intervals. When they lift slightly (Quartile #3), the S&P’s average 12-month total real return becomes negative; and when they rise significantly (Quartile #4), it plunges to a loss of more than 7%.
According to The Wall Street Journal (30 November), “Federal Reserve Chairman Jerome Powell said the central bank was prepared to quicken the pullback of its easy-money policies, opening the door to raising interest rates in the first half of next year as it grapples with inflation and a potential new virus wave that could exacerbate supply-chain disruptions.” If that doesn’t chasten today’s bulls, a final point should:
Stagflation has prevailed in the U.S. since June 2020. During these months the S&P’s 12-month real total returns have greatly exceeded not just 2.8% (the mean total return during periods of stagflation) but also 8.6% (the mean during all 12-month periods). If past is prologue, for how long can recent stellar returns last?
Peter Lynch, one of the most prominent funds managers of the 1990s, famously declared: “If you spend 13 minutes a year on economics, you’ve wasted ten minutes.” Similarly, Warren Buffett once averred: “If Fed Chairman Alan Greenspan were to whisper to me what his monetary policy was going to be over the next two years, it wouldn’t change one thing I do.”
Most of the time, most investors should do the same – that is, ignore CPI, GDP and long-term rates of interest (as well as the prognostications of “experts” regarding their future course). In isolation, whether they rise or fall during a given 12-month interval – and by how much – matters little: on average, the S&P 500 has generated positive total returns.
Yet rules of thumb have important exceptions – and today might provide one. The confluence of rising long-term rates and consumer price inflation, together with stagnating GDP, has tended to generate appreciable investment losses. Moreover, that’s not taking into consideration the Australian economy’s longstanding, serious and deepening weaknesses and current bogus boom, and Australian stocks’ ongoing earnings recession and anaemic dividends. In conclusion,
Investors should consider two key questions: during the next year and beyond, will stagflation persist? Will long-term rates rise appreciably? If the answer to these questions is “yes” and historical patterns reassert themselves, then American (and, by extension, Australian) markets’ returns will cease to beat the odds. If stagflation persists and rates rise, investors should prepare for volatile markets and poor returns – including significant losses.
Noel Playford says
Another gem Chris, as good as May’s post re Australia’s bogus boom. So clinical and valuable compared to the opinion fed to the masses by media and other conflicted commentators. On that point, I’d like to see more public emphasis on per capita growth in Australia rather than the political self-serving GDP as a measure of economic performance.
Advice of your latest post has been the most anticipated of all my emails for years. Very much appreciated.
Chris Leithner says
Many thanks for the kind words!
Best regards, Chris