Feature Stories | Mar 25 2020
History is littered with bear markets and recessions. We’re in a bear market and a recession is inevitable. Can history tell us how deep and long this bear market and recession will be?
-Bear markets typically double-dip
-Bear markets vary in cause and duration
-The starting point is fundamental
-The end point is conditional
By Greg Peel
First, let’s consider some charts.
The following charts, courtesy of The Chartist, track “ASX” price movements over past bear markets.
Note that for the purposes of comparison, pre-1980 charts represent a compilation of the then regional exchanges. The All Ordinaries was introduced in 1980 and the ASX was formed in 1987, amalgamating six state exchanges. In 2000 the ASX200 replaced the All Ords.
The most famous stock market crash of them all saw a couple of false-start relief rallies before the final V-bounce, which was arguably as spectacular as the initial fall, despite ushering in the Great Depression. Subsequent volatility from 1937 cannot realistically be considered a double-dip of the ’29 crash, given other problems were mounting.
The first OPEC oil shock sparked the 1970-71 crash, and again a failed relief rally is evident within the fall. There followed the Whitlam government, stagflation, Watergate, the Yom Kippur War and a decade of recession, book-ended by the second OPEC oil shock.
Clearly the first crash was only a curtain raiser for the second, as hyperinflation gripped the world, in the days before central banks had any mandate to control it.
The 1987 crash is the most cited as a comparison for today’s market, in terms of speed. The All Ords fell -5% on the Monday, October 19, and those of us there at the time said thank God that’s over. The All Ords then fell -25% on the Tuesday, and it thereafter took four months to find a bottom.
The so-called “mini-crash” of 1989 then followed, sparked by the collapse of the leveraged buyout market in the US, famously highlighted by Kohlberg, Kravis & Roberts buying out RJR Nabisco in the biggest LBO in history. With that out of the way, the Japanese stock market crashed in 1991, due to a collapse in the Japanese property market, and in 1990, we entered Keating’s recession we had to have.
Thanks to a surge in commodity prices in the period, it took until 1994 for the stock market to respond.
The Australian stock market barely blinked when the Tech Wreck hit US markets in 2000, having minimal exposure to the dotcom sector. But then came 9/11. The bounce out of that event proved ill-timed and we entered the “mild” recession of 2002-03.
The next chart was published by Business Insider in 2015.
It took from October 2007 until March 2009 for the GFC bear market to bottom out around -50% down. There was a false glimmer when the Fed forced JP Morgan to buy out Bear Stearns, but Lehman Bros eventually proved one underwater investment bank too many.
QE1 brought a pretty swift bounce before problems began appearing elsewhere in the world, culminating in the European credit crisis of 2011 and the first use of the word “Grexit”.
Double Dipping
The upshot of the above is bear markets are rarely truly over until the second wave. Although it must be said that there is in some cases a dubious connection between the two waves.
The ’29 crash brought the Great Depression. The subsequent bear market in ’37 reflected a world heading into war, which ultimately ended the Great Depression. While the two events might seem mutually exclusive, it was the Depression that led to the rise of fascism in Germany, and elsewhere.
The ’29 crash followed the excesses of the Roaring Twenties. The 1970 crash was due to an unforeseen oil shock. The impact led to a decade-long recession the stock market initially failed to anticipate.
The ‘87 crash echoed that of ’29 in being the result of excess, most notably in Australia by the rise of the Alan Bonds, Robert Holmes-a-Courts and Christopher Skases. In the US few lessons were learned, given subsequent collapse of the leveraged buyout market and the Savings & Loans crisis.
The US dotcom bubble was a classic case of herd mentality and FOMO (fear of missing out). Few understood the internet, or dotcoms, but clearly this was the future. Ultimately it was, but not before a blow-off top in infinite PE ratios. It is unclear how this may have subsequently played out, given it was 9/11 that guaranteed the recession.
That recession covered the period 2002-03 when another issue was playing out – SARS.
In the seventies, mortgage-backed securities were invented. In 2002, collateralised debt obligations were introduced. By 2006, so great was the demand for CDOs, sub-prime CDOs were introduced. In the scramble to cash in on the bonanza, US investment banks were levered up to 40 times.
The rest is history.
If few understood the internet back in the nineties, hardly anyone understood a sub-prime CDO, which explains why the GFC bear market took so long to play out. By the time Lehman went under, the subsequent “crash” was almost the end of it.
It was two years later when the reverberations of the GFC exposed the eurozone as a farce.
They say that history never repeats but it does rhyme.
The ’29 and ’87 crashes and the Tech Wreck were boom-bust events. The OPEC oil shock and 9/11 were “black swan” events – crises unforeseeable. I think we’d all agree coronavirus is a black swan. Goldman Sachs further divides bear markets into three categories.
The Nature of the Beast
History suggests there are three types of bear markets. Goldman Sachs calls these the structural, cyclical and event-driven bear markets.
A structural bear market is triggered by structural imbalances and financial bubbles. Very often there is a “price shock”, such as deflation, which follows.
Cyclical bear markets are typically a function of rising interest rates, impending recessions and falls in profits. They are a function of the economic cycle.
Event-driven bear markets are triggered by a one-off “shock” that does not lead to a domestic recession (such as war, oil price shock, emerging market crisis or technical market dislocation).
Yes, I know what you’re thinking. Clearly the coronavirus is a one-off shock but global recession is now inevitable. Goldman Sachs (in a report published March 9) applies caveats in this particular case, In short, this time it’s different.
For the record:
Structural bear markets on average see falls of -57%, last 42 months and take 111 months to get back to starting point in nominal terms (134 months in real terms);
Cyclical bear markets on average see falls of -31%, last 27 months and take 50 months to get back to starting point in nominal terms (73 months in real terms);
Event-driven bear markets on average see falls of -29%, last 9 months and recover within 15 months in nominal terms (71 months in real terms);
…Goldman informs us.
The main difference between a “standard” interest rate led cyclical bear market and an event-driven bear market is less the severity of the fall itself but more the speed of fall and the speed of recovery, Goldman points out. These both tend to be faster in an event-driven downturn. Indeed, event-driven bear markets, on average, reach a low in around half a year compared with over two years for a cyclical bear market and nearly four years for a structural bear market.
Also, event-driven bear markets are on average back to their starting point typically within a year, compared to four years for a cyclical bear market and nearly a decade for a structural bear market.
On March 9, Goldman Sachs suggested “this looks like an event driven bear market” (and note that on March 9 the S&P500 had not quite reached -20% down yet) but the analysts made three important caveats:
Despite comparisons with SARS, no event-driven bear market has ever been triggered by a virus. They all have different characteristics but have all been internally market-driven, and hence a monetary response has usually been effective.
The Fed cut its cash rate by -50 basis points on March 3 and by -100 points to zero on March 15 and is of now providing unlimited monetary support across asset classes.
In the case of coronavirus, Goldman Sachs questions whether monetary policy can be effective in an environment of fear in which consumers are forced to stay home.
Moreover, never has a bear market begun in a period when global interest rates are this low (and negative in some cases), providing less room for a monetary response.
The SARS experience suggests the market should rebound once the pace of growth of the daily case-count slows. This has already happened in China, and possibly (as of last weekend) in Italy, but elsewhere (eg US, Australia) the true case-count is as yet undetermined.
The Starting Point
In the past 90 years, since the creation of the S&P500 index, there have been 36 stock market corrections, notes Longview Economics. Note that a correction is technically defined as a -10% plus pullback and a bear market as -20% plus. Those 36 corrections range in extent from -10% to -86%.
Longview suggests those corrections were all caused by some combination (hence overlapping) of three factors:
A recession (36% of corrections). On average, recession-related corrections take longer to play out and are larger, on an average fall of -36%;
A macro shock (56% of corrections). Examples include WWII, the Asian currency crisis (1998) and the eurozone crisis (2011). Of these 18 pullbacks, 10 were not accompanied by a recession. Average fall -21%;
A rising cost of capital (70% of corrections). In 15 cases, the rising cost of capital triggered the recession and/or shock while in 10 cases the rising cost of capital was not due to shock or recession. Average fall -16.5%.
Clearly, a macro shock is underway, Longview notes, undoubtedly triggering a US/global recession. This puts it in the category with 25 past corrections, which have an average of -20%. Only 8 prior have been larger than -30%.
The question over the month of March has been “has the damage now been priced in?” Of course, that question was asked, and often answered in the affirmative, at -10%, -20% and -30%. Longview ponders whether now (-30%) the damage has been priced in or whether conditions are in place to push this one up the ranks of the large, recession/shock-related corrections of the past 90 years.
An answer to the question requires an assessment of four key factors, Longview suggests:
- The extent of economic and financial market excess ahead of the shock
- The speed of the health policy response (speed of virus containment and time taken to develop a vaccine)
- The degree and speed of fiscal and monetary policy response
- The depth and duration of the recession
Regarding the 8 corrections in the past 90 years larger than this one, Longview divides those into two groups.
Five are associated with large and prolonged economic contractions. Three of those were associated with the Great Depression, one was the OPEC oil shock recession of the seventies and the other was the GFC.
Three are associated with a milder recession and/or major shock. They were the start of WWII (no recession), a mild recession in 1968-70 and a -50% pullback related to 9/11 (shock) coming on top of the Tech Wreck (unwinding financial market excess).
Indeed, all of the major recession-related corrections followed long phases of rising financial and economic excess, all preceded, in some combination, by a large (a) credit boom, (b) house price bubble, (c) current account deficit and/or (d) corporate sector financing gap. The ensuing recession was therefore deepened and prolonged by the unwinding of those excesses.
I know, again, what you’re thinking. In late February, both the S&P500 and the ASX200 hit all-time highs.
But in the US case, Longview suggests, the prevailing macro environment was “considerably more benign” than those of the five major economic contractions above.
The corporate sector was running a small cash flow surplus. In each of “the five”, the corporate sector was running cash flow deficits. There is no sign of excess in the housing market, the current account is small, household and financial sectors have been deleveraging for a decade, and the banking sector has been recapitalised (albeit the corporate sector has been leveraging up through the cycle).
We can also throw in historically low unemployment.
China has larger economic excesses, Longview notes, most notably in housing, credit and construction.
In Australia, despite a new stock market high, the economy was already in trouble, as evidenced by further RBA rate cuts ahead of the virus crisis. Unemployment was nevertheless stable, the current account was in surplus and the house price correction had bottomed. Consumers were deleveraging and not spending. If Australia was already headed for recession, it would have been driven by a combination of drought, bushfires and the government’s blind determination to return the budget to surplus.
Since then things are looking a little less dire on the drought front, post-bushfire rebuilding has begun (and the bushfires are now the tourism industry’s long forgotten threat) and the government has switched from surplus obsession to massive fiscal stimulus.
The RBA has implemented “unconventional measures” for the first time in history.
Longview’s conclusion is that while the US is not without economic and financial market excess, overall measures of fundamental excess in the economy are low. The virus outbreak is therefore unlikely to trigger a phase of economic pain that’s both deep and long lasting, provided the authorities are relatively effective at bringing the virus under control, therefore limiting the time economic activity is curtailed by containment.
Hence this correction will likely most resemble those three above associated with a milder recession and/or major shock. The ultimate size of the correction will be limited by policy response (monetary and fiscal), and by evidence (a) the virus is being contained, (b) has a lower mortality rate than feared, and (c) is eventually mitigated by a vaccine (probably 12 months).
The Bottom
Stock markets discount future earnings, implying the stock market will turn before the economy does. When the virus outbreak began economists factored in one quarter of economic contraction, before moving to two (technical recession) and perhaps now on to three or maybe more, depending on the conditions noted above.
There is no definition of a “depression” other than a recession that goes on for a very long time.
As the charts published earlier in this article suggest, typically such events feature a bear market correction followed by another downturn at a later date. All bear markets – even this one, despite the speed, feature the odd snap-back rally that is quickly snuffed out by previously slow movers on the sell-side. The question is as to when will this correction stop, and will that be the bottom?
All agree the correction will end the day it is clearly evident the pace of growth in the virus case-count, particularly in the US, is slowing. Arguably it might end before that, on the back of massive stimulus, if bold first-movers believe a light may be at the end of the tunnel.
All agree that when that day comes, the relief rally will be fierce. But then all agree the world is headed into recession. They just don’t know how deeply and for how long.
If there is to be a second-wave correction following a sharp relief rally, it will likely be triggered by data-shock (not to mention an expectation a second wave must follow). We have not yet posted one quarter of contraction, let alone two.
February economic data are now misleading and March data will still be half misleading so we’ll have to wait until at least May to see the true impact on most data series. In April we’ll get a first estimate of March quarter US GDP, and it will not be until July a recession can officially be called, despite the fact everyone will know they’re in one by that point.
In Australia, we won’t get a March quarter GDP reading until June, and a June quarter until September.
One factor oft underlined is that this particular crisis is impacting on both the demand side (locked up consumers) and supply side (shuttered factories) of the economy, making it “unique”.
“Unique” is not encouraging, as it implies “this time it’s different”, and history suggests it never is. Yet looking at all the history outlined in this article, it is difficult to make a direct comparison with history.
Indeed, the closest comparison, if we think of lockdowns and food shortages, is the War, as many have suggested. During the War, falling stock market prices were the least of one’s problems.
Technical limitations
If you are reading this story through a third party distribution channel and you cannot see charts included, we apologise, but technical limitations are to blame.
Find out why FNArena subscribers like the service so much: "Your Feedback (Thank You)" – Warning this story contains unashamedly positive feedback on the service provided.
FNArena is proud about its track record and past achievements: Ten Years On
