Key takeaways:
- Stagflation seems rarer than the headlines suggest, but the conditions today bear watching. Of 27 US recessions since 1890, only seven meet a working definition of stagflation; and all have occurred around a war, a monetary policy regime change, or both. With tariff / war-driven inflation and questions over Fed independence, some of the ingredients are present, even if the data isn't there yet.
- Trend-following seems to be the most consistent stagflation hedge; gold was a surprise disappointment. Trend strategies worked in six of the seven episodes, with the sole exception being 1980 — when Volcker's abrupt regime change created whipsaw conditions. Gold, despite its reputation, delivered only in 1973–75, when its performance was driven by the exceptional unwinding of the Bretton Woods peg.
- Equities draw down but have usually recovered, and bonds depend on monetary credibility. Historically, stocks fell in every stagflation bar the first, but declines were often shallower than feared and recoveries swift. Bonds performed well when central bank independence was firmly asserted (1957–58, 1981–82), and badly when it wasn't (1918–19, 1973–75).
There's no agreed upon definition as to when a Baroque nude becomes adult entertainment but, as the adage goes, you'll know it when you see it. Similarly, ever since the British Conservative politician Iain Macleod coined the term ‘stagflation’ in a speech to parliament in 1965,1 there has never been an empirical threshold established. Yet people have seen it everywhere. Today, it’s more explicit than ever, with data from Google Trends suggesting a peak in interest in March, in the wake of the stagflationary implications of the Middle East war, and not far off the zenith of June 2022.
But what is it? I’m no expert (the British people have had enough of them), but here’s a working definition which likely does the job just as well as a long paper with lots of scientific notation: a National Bureau of Economic Research (NBER) recession where inflation annualises at 3% (100 basis points [bps] above the Federal Reserve’s [Fed] target).
In Figure 1, I show annualised headline inflation for all NBER recessions back to 1890. Of the 27 America has experienced over the last 150 years, eight (30%) are outright deflationary, 12 (44%) see low but positive inflation, and seven (26%) are stagflationary, on my definition (Figure 1 shows Deflation below 0% in aqua blue, Low Inflation 0-3% in royal blue and Inflationary as more than 3% in fuchsia). It’s unusual, in other words. So I wouldn’t want to say much with huge conviction, but for what it’s worth, here’s my whistle-stop on the seven. For each, I leave my reader to decide whether it’s the Rokeby Venus, or something seedier.
Figure 1: Annualised inflation through NBER recessions
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: January 1890 - December 2020.
1918-19: The Great War ends
It’s often said that 1900 was a bad time to be born, and this stagflationary period was in the thick of it, with the horror of war ending but the baton being passed to the Spanish Flu, killing at least 50 million people worldwide by some estimates.2 The Armistice of November 1918 sparked a rapid unwind of the total mobilisation command economy that most of the world had been operating under for four years. The labour market was dislocated not just by the pandemic, but by millions of demobilised soldiers in need of reabsorption. Meanwhile, wartime price controls were released, and the kinetic energy of long pressured prices came on full display. At the same time, the Fed, still in effect a sub-department of the Treasury prioritised orderly bond issuance markets over consumer price stability, dovishness which came on top of a large increase in the monetary base through the war. So, while the labour market was in the midst of an earthquake, there was still too much liquidity chasing too few goods.
Figure 2 shows real annualised returns to equities, bonds, gold and commodities through the episode, when annualised inflation was 11.4%. Not the end of the world, the market bottom line. Commodities were the only asset that really experienced any kind of precipitous drawdown, and even then, only just breached the bear threshold (-20%) and recovered substantially into stagflation’s end. Stocks were flat to slightly up throughout. Gold and bonds were the losers, but only in line with inflation. That makes sense given a) the US was on a gold standard at the time so the yellow metal was fixed, while inflation was not, and b) bonds were in a semi-managed market, so again, in real terms, I’d expect a detraction roughly of the magnitude of the CPI rise.
Figure 2: Performance through Stagflation One
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 31 August 1918 – 31 March 1919.
1957-58: A poor man’s stagflation
This one only scrapes in, per our definition in Figure 1. But a threshold is a threshold, and what is more, it is one of only two stagflation episodes where CPI year-over-year accelerated (the other being 1973-75, which we’ll come to). On the ‘stag’ side, the proximate trigger was a tightening cycle led by Fed Chair William McChesney Martin, often seen as the architect of US central bank independence, much to the chagrin of various presidents who had assumed he would be a monetary valet. At the same time the economy was experiencing an inventory correction, following a capital goods boom in the mid-1950s. On the ‘flation’ front, this was the episode that birthed the phrase ‘cost-push inflation’ into the economics textbooks, a cocktail of administered cost-plus pricing in concentrated, heavy-manufacturing industries, and strong union wage bargaining power.
By this point, we have more comprehensive asset return data, shown in Figure 3, when annualised inflation was 3.2%. This time, bonds were the outperformer, trading freely, and in an environment where central bank independence was being muscularly asserted, as previously discussed. This was a boon both for the term premium and inflation expectation components of fixed income returns. On the negative side, stocks were the clear loser, although again, the magnitude of declines was limited, and on a monthly periodicity basis, well shy of a bear market, even in real terms. Gold was not the hedge I would expect; again, this is a mechanical effect from it still being fixed to cash. Looking across factors and strategies: Value – which likes inflation but dislikes recession, or so the market folklore goes – seems to be reacting more toward the latter. Trend and Momentum working suggest the themes of this stagflation were present in its build-up.
Figure 3: Performance through Stagflation Two
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 30 August 1957 – 30 April 1958.
1969-70: Guns and butter
Though relatively mild, both in terms of depths of the growth drawdown, and the heights of the inflation heat-up, the 1969-70 recession contained many of the best hits that would be played on repeat through the 1970s. In the mid to late 1960s, the Johnson administration had simultaneously pursued policies of escalation in Vietnam (‘guns’) and the Great Society programme of welfare reform (‘butter’). The resulting fiscal excess sent the chickens home in the form of uncomfortably hot inflation, even as the Fed tightened policy through 1969. Moreover, the first crack in Bretton Woods had come in March 1968, with the introduction of the two-tier gold market, limiting dollar-gold convertibility to central banks. Implicitly, this was hat tipping the reality of the Triffin Dilemma: that global liquidity requirements had reached a level where an explicit one-item peg was becoming untenable. This was likely inflationary in terms of eroding dollar purchasing power.
This inflation came hand-in-hand with a credit crisis triggered by the Fed’s raising of rates, culminating in the bankruptcy of Penn Central in June 1970, at the time the largest corporate insolvency in US history.3 Such tightening of financial conditions was also seen in the equity market, through the 1968-70 bear market, and the unwind of a bout of stock speculation, with a technology and small-cap flavour to it.
In this context, the results of Figure 4 are intuitive (annualised inflation 5.6%). Stocks sold off, in more pronounced fashion than our first two stagflations, and with an emphasis on the Size factor (in other words: small underperformed large-cap). On the other hand, hard (or harder) strategies outperformed (commodities and Value), and gold saw its first positive stagflation, in real terms, being allowed to move at least partially freely. Meanwhile, in an early sign of the flight to so-called ‘one decision stocks’ that would come to define the Nifty Fifty bubble that inflated over the next three years, Quality was also a strong performer.
Figure 4: Performance through Stagflation Three
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 31 December 1969 – 30 November 1970
1973-75: The big one
Central casting’s stagflation was a perfect storm of economic malaise. The OPEC oil embargo quadrupling oil prices, the final collapse of Bretton Woods, the unwind of the Nixon wage-price controls, the fiscal capture of Arthur Burns, and a number of global harvest failures to boot, gave rise to a stagflation like no other, at least since the NBER series begins, and certainly not since. Industrial production fell 15% peak to trough, inflation annualised close to 11%, and the rate of CPI year-over-year markedly accelerated by 200 bps (from 8% to 10%).
In this context, gold struck an innings that would make its reputation as an inflation, and particularly stagflation, hedge, with a 55% enhancement to purchasing power, per Figure 5 (annualised inflation 10.8%). For me, there’s something a bit ‘one hit wondery’ about this, but I’ll come back to that. We again see the rewarding of hard assets, or hard-asset oriented strategies (commodities and Value both performed well). We also see Trend exhibiting strong returns, implying the patterns of this extreme stagflation were built with relatively few breaks. Equities feel the sharp end of the pain, but this time bonds are not spared. Like our first episode, and unlike the interceding two, this stagflation was accompanied by limp monetary policy. Finally, it is worth noting that, although the opening stages of the sell-off were heavily influenced by the Nifty Fifty crash, Quality as a long-short factor, while negative, was not disastrous. This implies that many of these names were as much caught up in massive macroeconomic forces, rather than falling victim to a style rotation in isolation.
Figure 5: Performance through Stagflation Four
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 30 November 1973 – 31 March 1975.
1980: The rise of Volcker
The Islamic Revolution of 1978-79 took around 5% of global oil supply off the market. This coming on top of the inflation expectations embedded earlier in the decade (for instance in the preceding stagflation episode), and the Fed chairmanship of G. William Miller – Arthur Burns’ equally ineffectual (and short-lived) replacement – as well as the Carter administration’s misguided attempt to control inflation via credit controls, caused fast but profound stagflation. The ultimate result was the determination of Paul Volcker, appointed in August of 1979 and the modern central banker’s pin-up, to prioritise price stability over the labour market.
Looking at asset returns (Figure 6, annualised inflation 13.8%), the most striking difference to the charts so far is the level of reversals through the episode. Most notably in gold, at one point up 25%, at another down 8%, before being up 19% again. Such whipsaw makes sense in the context of a market digesting Volcker’s moment of truth in enacting a fundamental monetary regime change. Would he chuck the ring into Mount Doom, or fail like Isildur? With the beauty of hindsight, the ‘Great Moderation’ can seem like an inevitability, but at the time it would not have felt that way, and Volcker himself arguably wobbled in the lead-up to the 1980 election. Between the sunlit uplands of secular stagnation and gaping jaws of stagflation lay a gulf of uncertainty. Here the patterns were inconsistent, and Trend consequently suffered.
Figure 6: Performance through Stagflation Five
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 31 January 1980 – 31 July 1980.
1981-82: Volcker triumphant
Or, ‘The End of Stagflation’. The recession of July 1981 to November 1982 was deliberately engineered to finally slay the inflationary dragon. Volcker took rates to unprecedented levels in maintaining a monetary tightening under bone-crunching political and popular pressure. While it gets past our stagflation threshold (see Figure 1), it does so via residual inflation, which meaningfully decelerated through the episode.
Looking at the asset returns in Figure 7 (annualised inflation 5.7%), we see the re-establishment of monetary credibility in the strong performance of bonds, despite the pronounced hiking, as the market priced the end of inflation and allowed fixed income to inhabit its traditional ‘flight-to-safety’ role in a recession. Moreover, gold was negative, as fiat finally flexed its muscles. The other pain point was broader commodities, intuitive in that this was a recession triggered on purpose to destroy demand for stuff, and it worked. Equity declines were very contained for such a long recession, again likely relief pricing of policy stability. Value is perhaps the surprise, and is a reminder that, historically, the factor’s sweet spot has generally been disinflation environments coming off high levels, as opposed to simple inflation.
Figure 7: Performance through Stagflation Six
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 31 July 1981 – 30 November 1982.
1990-91: The stagflation that wasn’t
This recession, now largely forgotten in the popular memory (although probably not by George Bush Sr), was triggered by the oil price spike which came out of Iraq’s invasion of Kuwait in August 1990. This was the fuse on a tinderbox that contained the long and variable lags of new Fed Chair Greenspan’s first tightening cycle of 1988-89, the aftereffects of the savings and loan crisis and the interlinked commercial real estate collapse. It was the first recession post the 1970s inflationary concern, and though the immediate catalyst was an energy market overheat, the fire didn’t make its way through to core CPI measures in the same way as prior episodes, and it therefore gets into our list on something of a technicality.
The price action (Figure 8, annualised inflation 5.3%) will now be relatively familiar. Pattern-seeking strategies (Trend and Momentum) worked. Quality worked. Gold did not work. Commodities spiked initially on oil, and then sold off through the rest of the episode. Equities had a noticeable if manageable drawdown, followed by a pronounced recovery. Value was the laggard, perhaps an early precursor to Graham & Dodd being a paradigm for failure through the rest of the decade, as tech-euphoria was unleashed.
Figure 8: Performance through Stagflation Seven
Problems loading this infographic? - Please click here
Past performance is not indicative of future results. This analysis is based on our research and is intended for illustrative use based on considerations listed in this paper and should not be construed as a recommendation and should not be relied on. Source: Bloomberg and Man Group. Date range: 31 July 1990 – 31 March 1991.
Seven lessons from seven stagflations
Was that all just intellectual masquerading? Here are seven lessons which, I hope, give it some point. First, true stagflations are rare. You could well argue that three of our seven (the most recent two and the 1957-58 episode) are not the real McCoy, which would give you just four episodes in 150 years, with nothing in the last 45. Second, bonds have worked in stagflations that coincide with monetary policy credibility (1957-58, 1969-70, 1981-82, 1990-91), and failed in the opposite (1918-19, 1973-75). Third, equities tend to draw down (doing so in every episode apart from our first one). But the magnitude of this trough is often less than you might have expected, and the recovery is usually pronounced, and even within the episode itself. Fourth, Trend has worked in every event bar 1980. On this showing, it is one of the best stagflation hedges there is, with the one caveat that (for obvious reasons) a regime volte-face is not its friend. I would argue 1980 was a prime example of that. Fifth, commodities tend to do better in a stagflation environment than a conventional recession, on average +5.5% real annualised, compared with -0.6% in all recessions. But, sixth, gold was a disappointment. So the bugs tell us: it has both ‘crisis alpha’, and is a fiat alternative. The combination of the two should be deep in the wheelhouse. It’s not. Seventh, historic stagflations have always come in and around a war, a monetary policy event, or some combination of the two. Are we there today? War, tick. Monetary tectonics, jury’s out. In one telling, the Fed’s independence is threatened in a way that it hasn’t been for 50 years. Stagflation is not borne out in the numbers yet, at least not on the growth side of the equation. But some of the pieces are there. On 10 March 1914, Mary Richardson took a meat cleaver to the Rokeby Venus.4 Forewarned is forearmed, and hopefully this little stagflation survey will mean that you, dear reader, should the real thing appear, will react less violently.
1. Apparently, there’s some speculation that American economist Paul Samuelson was the true originator. Patriotic fellow that I am, I’m going with Macleod
2. Centers for Disease Control and Prevention. (11 May 2018). 1918 pandemic (H1N1 virus) https://archive.cdc.gov/www_cdc_gov/flu/pandemic-resources/1918-pandemic-h1n1.html
3. "The Financial Collapse of the Penn Central Company" — House Subcommittee Staff Report, 1972 https://fraser.stlouisfed.org/files/docs/historical/house/1972house_fincolpenncentral.pdf
4. The Manchester Guardian. (11 March 1914). Suffragette outrage. Rokeby Venus slashed with a chopper. Sequel to Mrs. Pankhurst's rearrest.
You are now leaving Man Group’s website
You are leaving Man Group’s website and entering a third-party website that is not controlled, maintained, or monitored by Man Group. Man Group is not responsible for the content or availability of the third-party website. By leaving Man Group’s website, you will be subject to the third-party website’s terms, policies and/or notices, including those related to privacy and security, as applicable.