Rising recession risk
For several months, we have been flagging the significant and rising risk of recession. In addition to the pre-existing supply chain and inflation morass, the combination of increasingly aggressive central banks, a stumbling China and the commodity shock emanating from the war in Ukraine have naturally further dimmed the growth outlook. The high level of uncertainty surrounding many of those and other variables highlights the possibility of scenarios sharply worse (though also considerably better) than the base-case outlook.
Over the next 12 months, we continue to flag a recession risk for North America of around 30%. The European risk is somewhat higher given a greater exposure to Ukraine – around 40%. These probabilities are three to four times higher than usual, although it must be emphasized that they nevertheless mean that a continued economic expansion is the most likely outcome for the coming year.
But it is important to think beyond merely the next 12 months. The risk of recession at some point over the next two years may well be above 50%, and potentially by a significant margin.
Why so high? Monetary tightening should theoretically exert the greatest weight on 2023 rather than 2022, for two main reasons:
- Much of the tightening won’t happen until 2022 is significantly underway.
- Monetary tightening operates with a lag on the economy.
Reflecting this, a 2yr-10yr yield curve usually spells trouble in the second year after the inversion, not the first. The business cycle will also likely have advanced further by then, and the tailwinds from the pandemic recovery should be all-but complete by then.
The current round of monetary tightening deserves some extra consideration. Even under normal circumstances, it is unusual for an economic expansion to survive a tightening cycle. Indeed, by one count, 8 of the 11 U.S. tightening cycles since World War II have ended in recession. So the odds aren’t good to start with.
But they are arguably worsened by the fact that this particular tightening cycle started with a policy error: the tightening should have begun in 2021. The implication is that the pace of tightening must now be unusually fierce and rates must venture quite far from their starting point. More broadly, it is not ideal that the main motivation for monetary tightening is excessive inflation rather than excessive growth. In particular, it means that, should the economy start to falter, central banks won’t necessarily halt their tightening because their main priority will be to wrestle inflation to the ground. Put differently, the Fed won’t step in this time, nor will other central banks.
The simultaneous arrival of a commodity shock doesn’t help the recession outlook. Even when ignoring the increase in natural gas and food prices spurred by the war in Ukraine, the oil shock has been sufficiently large that a recession results more often than not from a leap of this magnitude (see next chart). It is fair to point out that oil isn’t as central to economic growth as in the past, such that this relationship may be exaggerated in the modern context, but it isn’t helpful.
Oil price shocks of more than 50% typically precede or coincide with recession
As of 04/20/2022. Shaded area represents U.S. recession. Source: Macrobond, RBC GAM
It might seem easy enough to dial the growth rate down to a sufficiently modest level that supply chain and inflation pressures ease as the unemployment rate edges away from its current rarified reading. But in practice it isn’t. Such soft landings are made especially challenging by the fact that economies become quite slippery once they have overheated. Going back to World War II, a recession has resulted every time the three-month average of the U.S. unemployment rate rises by a mere 0.4ppt or more (see next chart). So there isn’t an obvious path to take the U.S. unemployment rate from the current white hot reading of 3.5% back to a tamer reading of, say, 4.5% in an effort to extend the cycle.
Important not to overheat U.S. labour market
As of Mar 2022. Unemployment rate is 3-month moving average. Source: Bureau of Labor Statistics, National Bureau of Economic Research (NBER), Haver Analytics, RBC GAM
On the other hand, and in an effort to highlight that the risk of a recession over the next two years remains well short of 100%, there are plenty of ways the expansion could persist.
It may be that only a slight economic deceleration is needed to take the pressure off supply chains and commodity prices, with inflation beginning to fade after a surprisingly small amount of monetary tightening. An easing of supply chain problems could also unleash a new economic tailwind as pent-up demand for previously unavailable goods is filled.
The pandemic could effectively vanish, much as eventually happened with the Spanish Flu, unleashing more Chinese growth in particular. Fiscal policy may also come to the rescue with greater force than usual – as it did to limit the damage during the pandemic.
Inflation might abate more quickly than expected as a fraction of prior price increases reverse as shipping costs normalize, car prices settle down and so on.
If interest rates are deflated by the current rate of inflation – yielding a version of a real interest rate – current borrowing costs are extraordinarily negative and have fallen to 70-year lows even as central banks tighten rates (see next chart). This is increasingly stimulative, not restrictive.
Real yields have been falling since the onset of the pandemic
As of 04/21/2022. Measured as nominal 10-year Treasury yield minus year-over-year percent change in Consumer Price Index (CPI). Shaded area represents recession. Source: Macrobond, RBC GAM
We discussed at some length the various recession claims being made by different parts of the U.S. yield curve a month ago, here. In a nutshell, the 2yr-10yr part of the curve claims recession by flirting with a negative spread, while the 3m-10yr does not. At the same time, the Fed’s favorite part of the shortest end of the curve argues even more strongly that no recession is coming (see next chart).
Mixed message on business cycle from yield curve indicators
As of 04/21/2022. Near-term forward spread measured as forward rate of 3-month Treasury bill six quarters from now minus spot 3-month Treasury yield. Shaded area represents recession. Source: Engstrom and Sharpe (2018). FEDS Notes. Washington: Board of Governors of the Federal Reserve System, Bloomberg, Haver Analytics, RBC GAM
However, it should be noted that these metrics may not be quite as different as they first appear. In the lead up to a recession, the 2yr-10yr spread normally inverts first, followed by the 3m-10yr and finally the Fed’s measure. This makes sense given what each is measuring. If the Fed delivers as much tightening as seems likely over the coming year, the 3m-10yr yield curve could easily be inverted by the end of the period. And, indeed, history shows that the 2yr-10yr inversion generally precedes a recession by a long 18 months, while the – as yet theoretical – inversion of the 3m-10yr curve usually doesn’t happen until closer to a recession.
Framed differently, the best-case scenario over the next two years is that economic growth persists and inflation is tamed – a soft landing. The worst-case scenario – and a fairly unlikely one – is that growth falters while high inflation remains. But what about the other two scenarios in this two-by-two matrix of growth versus inflation?
Normally, one might imagine that the scenario in which growth persists and inflation remains high would be the second-best outcome. But that’s probably not the case right now. The second-best outcome is probably a recession in which inflation is successfully tamed.
It is tempting to conclude that if the second-best outcome is a recession, all hope should be lost. But it may well be the opposite: risk assets have been fairly resilient as additional monetary tightening has been priced in. This is presumably because markets recognize that taming inflation is far more important for long-term prosperity than squeezing out another year of economic growth. It may well be that a “successful” recession is treated with calm by financial markets. Further, any such hypothetical recession is some distance off, and risk assets often preform just fine in the first year after monetary tightening is undertaken.
Pandemic in China
Although a BA.2 wave of some significance continues to ricochet around the world, it has proven to be only minimally consequential for ex-China economic growth. The wave has already crested and receded in Europe. Canada may also be starting to improve. The U.S. is not yet, as the number of states experiencing a rising infection count is still rising (see next chart). But it is also likely to peak before too long and to do minimal economic damage.
Number of U.S. states with transmission rate above key threshold of 1
As of 04/22/2022. Transmission rate calculated as 7-day change of underlying 5-day moving average of new daily cases, smoothed with 7-day moving average. Transmission rate above 1 suggests increasing new daily cases. Includes Washington, D.C. Source: Haver Analytics, Macrobond, RBC GAM
The exception to this critique is China, which struggles to maintain its zero tolerance policy against a highly contagious variant. Now Beijing is beginning to lock down in a similar manner to Shanghai. This does significant domestic and international economic damage and is a major reason why markets have been sour in recent days. It is also why we continue to downgrade our 2022 Chinese growth outlook – now to just 4.4% – well below the country’s own 5.5% growth target. It is hard to fathom that China will be able to put the lid back on COVID-19 from here.
While addressing China in the context of the pandemic, let us solve two mysteries. The first mystery is why more than 90% of China’s cases were deemed asymptomatic when the global asymptomatic rate is thought to be no more than about 25%. Are infected people concealing their status so as not to be subjected to China’s harsh rules, or might China be intentionally undercounting? Not exactly. Instead, China’s definition of “symptomatic” is simply strange. China only deems a patient symptomatic if they have had a lung scan that demonstrates infection. Naturally, few people receive such a scan. So it is best to ignore the asymptomatic vs symptomatic distinction, and it is worth seeking out the asymptomatic numbers which can be harder to find.
The second mystery is why China’s fatality rate is so low. At one point, China had reported 443,000 new infections since March 1, versus just two deaths. There were theoretically no deaths in the hotbed of Shanghai for the longest time. Clearly that can’t be right. And indeed it isn’t. If someone who dies has any underlying condition, China reports the cause of death as the underlying condition. Thus, there has been a spike in people dying of diabetes and heart disease. China isn’t totally wrong to the extent those people probably wouldn’t have died without those underlying conditions. But it still seems much more logical to record COVID-19 as the cause of those deaths. As such, China’s reported fatality numbers are not helpful.
Ukraine peace prospects fade
Russia continues its major push in eastern Ukraine, with the next few weeks set to be pivotal.
Ukraine may have damaged another Russian oil depot on Russian soil.
The war is likely to remain focused on eastern Ukraine for the moment and perhaps for good. Yet a Russian general expressed a desire to also take southern Ukraine and neighbouring Moldova’s long-disputed Transnistria region – which has already had Russian troops on its soil for years.
Ceasefire prospects continue to fizzle. In mid-March the likelihood that the conflict would endure past December 1 was deemed to be just 13% according to an online betting market. That figure has now soared to a 62% chance. We assume this will be a multi-year war, and in turn that sanctions will persist.
On that note, we continue to flag the risk that international sanctions rise, especially with regard to Russian energy. Sixty percent of large international companies surveyed by the Yale School of Management have now withdrawn from or suspended their operations in Russia.
Furthermore, despite somewhat contradictory comments from politicians, sanctions are unlikely to be lifted any time soon. Not only does the war appear set to last for a considerable period of time, but it is unlikely that most sanctions would be lifted even if the war ended. A study of historical sanctions finds that the average sanction lasts a whopping 16 years (though the inclusion of permanently sanctioned countries like Cuba inflate the estimate). The key point is that sanctions are usually measured in years rather than months.
NATO membership is about to grow in response to the Russian threat. Finland and Sweden are both reported to have prepared NATO applications, setting them up for lightning-fast admission at a June NATO summit.
Despite the narrowing of the war in Ukraine, there has not been a material return of refugees or displaced people. To the contrary, the number of both has increased by a further 2 million people in recent weeks to reach 12 million out of a population of 44 million.
Notwithstanding the elevated risk of recession over the next few years, global economic momentum has been decent to good in recent months. Economic surprises have actually turned back into positive territory (see next chart).
Global economic surprises rebound
As of 04/22/2022. Source: Citigroup, Bloomberg, RBC GAM
While the recent tightening of financial conditions is serious, the magnitude of the move is on par with a mid-cycle slowdown (such as 2015—2016) rather than a recession (such as 2008 or 2020 – see next chart).
Global financial conditions tighten quickly
As of 04/20/2022. Source: Goldman Sachs, Bloomberg, RBC GAM
News sentiment – while still negative – has become incrementally less so in recent weeks (see next chart). Whether this is because conditions have genuinely improved or because journalists are moving on to fresher topics is unclear. But it may not matter to the extent households and businesses find themselves less beaten down by the latest barrage of news.
Daily news sentiment in the time of COVID-19
As of 04/17/2022. Source: Federal Reserve Bank of San Francisco, Macrobond, RBC GAM
The near-term economic outlook will be significantly influenced by the consumer. On that front, there are an unusually large number of negative signals, but also many positive signals (see next graphic). We conclude that consumer spending growth can be decent over the next few quarters, but with an unusually wide range of conceivable outcomes.
Tricky consumer outlook – decent spending most likely, but uncertain
As at 04/22/2022. Source: RBC GAM
At a minimum, the upside to consumer spending isn’t as large as it once was now that the U.S. personal savings rate has slipped from extremely elevated levels to a slightly below-average reading (see next chart).
U.S. personal saving rate slips below pre-pandemic levels
As of Feb 2022. Shaded area represents recession. Source: U.S. Bureau of Economic Analysis (BEA), Macrobond, RBC GAM
Key inflation ideas
There are three main messages to share about inflation and then a few smaller points.
Inflation is high
First, and to state the obvious, inflation remains very high, and indeed continues to rise. U.S. Consumer Price Index (CPI) is now a whopping +8.5% year-over-year (YoY) and in Canada the rate has now reached +6.7% YoY. Inflation surprises continue to be almost uniformly above the consensus (see next chart). This is the side of consensus our own forecasts have stuck to for the past year and we remain there now.
Global inflation continues to exceed expectations
As of Mar 2022. Source: Citigroup, Bloomberg, RBC GAM
Inflation has broadened
Second, inflation continues to broaden as more and more product prices accelerate. What was originally a problem with computer chips and used cars and a small handful of other items has now infected well over half of the CPI. More than half of the components in the U.S. CPI are now rising by more than 4% per year (see next chart). Very few are rising by less than 2%.
Rising inflationary pressure in the U.S.
As of Mar 2022. Share of CPI components with year-over-year % change falling within the ranges specified. Source: Haver Analytics, RBC GAM
This breadth will make it much harder to tame the inflation going forward, and largely explains the urgency central banks suddenly feel.
Long-term inflation may be a bit higher
Third, we are increasingly of the mind that long-term inflation may end up being a bit higher than it was before the pandemic. Whereas, stylistically, inflation trended toward something like 1.5% before the pandemic, it might trend toward something like 2.5% afterward. In both cases, this is fairly close to the 2.0% target maintained by most central banks. It is also a perfectly reasonable level for inflation: it is within the broader 1—3% comfort zone and at a level that is historically associated with perfectly normal economic growth and adequate market returns.
Some prominent economists – including the former chief economist of the International Monetary Fund (IMF) – have advocated for a higher inflation target because it provides more room to deliver monetary stimulus during times of trouble. However, nominal interest rates would have to be a bit higher than otherwise in that environment, and plausibly central banks might also maintain a slightly higher real rate as they try to exert a slight downward pressure on inflation. The effective tax rate on investment income also rises marginally in a higher inflation environment.
The arguments that we could be transitioning into a slightly higher inflation environment over the long run are as follows:
- Inflation expectations may prove sticky at a slightly higher level after this bout with extremely high inflation.
- De-globalization is intensifying – not only was a degree of trade saturation seemingly reached around a decade ago, but several other factors have come into play. Antagonism between China and the west grew under the Trump administration, supply chain resilience has become a priority during the pandemic, and now the world further decouples as Russian sanctions mount. All of this adds up to a pinch of additional inflation per year.
- Climate change continues and mitigation efforts are accelerating. These are broadly inflationary, including the cost of rising carbon taxes, the cost of replacing a dated capital stock, the cost of natural disasters, and the possibility that the supply of fossil fuels will fall more quickly on concerns about stranded assets than does the demand for them (raising their price).
- Lastly, we believe workers are gaining additional clout for a myriad of reasons ranging from fast-rising minimum wages to greater attention from policymakers to a rising dependency ratio, whereas corporations are losing clout as global minimum corporate taxes arrive and anti-trust investigations strengthen. All of this may add up to more wage growth and more inflation. Productivity-adjusted labour costs began to tilt in a less deflationary direction a full decade ago (see next chart).
U.S. productivity-adjusted labour cost has been rising since start of pandemic
As of Q3 2021. Shaded area represents recession. Source: BLS, Haver Analytics, RBC GAM
To be clear, demographics should still be profoundly deflationary, but no more so than it has already been since the late 2000s.
We now turn to some smaller inflation thoughts.
No inflation below 2% this cycle
First, inflation seems unlikely to descend below 2% before the end of this cycle. Either a recession happens, in which case the cycle is over before inflation can descend too far, or growth persists with a very tight economy (recall the earlier chart detailing that it is rare for the unemployment rate to rise materially without a recession occurring) and inflation remains higher than normal, if hopefully well below current levels.
Any inflation give-back?
Second, should we expect inflation to run at a substantially lower-than-normal rate as prior large price increases are around?
On the one hand, it is hard to think of a reason why shipping costs have to remain permanently high after supply chain issues have been resolved. Similarly, it would be strange if used car prices remained permanently higher than before once the supply shortage is addressed. Food prices can probably come back down as growers adjust to Russian sanctions. All of this argues that there could be a period of unusually weak inflation or even deflation as these issues are resolved.
On the other hand, there wasn’t a decade of low inflation after the 1970s: price increases can very easily prove permanent. Further, central banks are not price-level targeters: all their mandates ask is that they return inflation to a normal rate, not unwind earlier high inflation. A period of deflation could be problematic as it would necessitate quite high real interest rates – a potential killer of economic growth. Lastly, to the extent that wage growth is picking up, and that wages are resistant to decline, it would be reasonable to think that at least part of the price increases will prove permanent if the labour portion of the producer cost base has increased.
These contradictions can be resolved in that specific acute distortions will probably be unwound at a time when broad inflation expectations are still somewhat elevated, with the two forces pushing in the opposite direction. Inflation might be inclined to run at 5% for a period of time if expectations had their way, but will instead run at 3% over that span as certain relative price movements unwind.
Central banks can’t fix inflation by themselves
We begin with the assertion that central banks did not cause high inflation all by themselves. Economies running slightly above their economic potential do not normally spit off inflation rates that are 6 percentage points higher than normal. A 2.5% or 3.0% rate would be more logical in that environment.
The roots of today’s high inflation are mainly the changing composition of demand and certain supply-side pressures including Russian sanctions. The persistence of those pressures have since permitted inflation to broaden out to most products.
In the same vein, central banks cannot fix high inflation all by themselves. A simple rule of thumb in which each 1 percentage point of rate hikes lowers inflation by a third of a percentage point would require a startling 18 percentage points of monetary tightening to bring inflation back to the realm of 2%. Yes, the Volcker Fed raised rates to approximately that level, but not from such a low starting point, nor in the same broader economic or inflation context. It is ultimately an unreasonable expectation.
Yes, central banks might expect to have an outsized influence on inflation expectations at a pivotal moment like this, but the basic point remains.
For inflation to come back down, some of the other issues are also going to have to be fixed. It is not unreasonable to think that supply side problems may abate naturally as demand organically moves back toward services. A mild economic deceleration could also have an outsized effect on supply chains. Similarly, while there remains some upside risk to commodity prices all that is needed is for commodity prices to rise less quickly over the coming year than they did over the last year – not to fall back to prior levels. Fingers crossed that these things happen, or central banks may have a very long way to go.
The “Great Resignation”
The “Great Resignation” is a term conventionally used to describe the various ways labour markets have been distorted by the pandemic. It is most commonly used to describe either the reduction in the labour force participation rate in the early stages of the pandemic as a cohort of people stopped working (or looking for work), or the fact that a lot of people are switching jobs right now.
In actual fact, it may be best to step back and consider the way that labour markets have changed (and may change) over four different time periods.
- Over the past decade, the rate of job-quitting was already rising subtly as the large cohort of baby boomers retired and as job tenure structurally declined over time.
- During the early part of the pandemic, there were many layoffs. Others quit to avoid getting sick, due to a lack of child care options, and/or were dis-incented from working by generous jobless supports.
- During the pandemic recovery, rapid hiring occurred, pulling most former workers back. But some remained on the outside, in many cases due to altered priorities or early retirement. Meanwhile, others left existing jobs in pursuit of greener pastures, for several reasons:
- The hot job market made it possible to move up more easily.
- People made strategic decisions to exit structurally depressed sectors.
- Pandemic-induced automation pushed workers out of some sectors.
- “High contact” jobs became much less attractive.
- Many people moved to new regions (see next chart).
U.S. job openings and quits hover around record high
As of Feb 2022. Estimates for all private non-farm establishments. Shaded area represents recession. Source: BLS, Macrobond, RBC GAM
This is arguably the most important phase of the great resignation because it is where we presently reside. The biggest takeaway is that by virtue of some people remaining outside the labour market and others opportunistically job hopping more than normal, wage growth and inflation are higher than they would otherwise be for the present level of employment.
- Over the long run, there should be some labour market scarring from people pushed out of the labour market by the pandemic who have since experienced a significant decay in their skills as they have sat idle. But, of greater importance, the potential for remote work is now very bright. With the ability to transcend geography and achieve a better work-life balance, it seems quite reasonable to expect the labour force participation rate to eventually be higher than it was before the pandemic.
Central banks continue to raise rates aggressively. The Bank of Canada provides a recent example of this, lifting its policy rate from 0.5% to 1.0% in a single swoop. Based on recent comments from the central bank, an even larger jump could be in store for the next meeting: do not underestimate the desire for central banks to reach a 2%+ level for short-term rates as quickly as is possible. The U.S. Federal Reserve’s own decision approaches in early May and seems likely to net a 50 basis point increase. Rates are rapidly moving off the floor.
A worrying (lack of) development
A worrying development in the monetary policy world is that inflation expectations have not come down even as rate hike expectations have grown. If markets thought the prospective tightening would be enough to tame inflation, inflation expectations should be settling back down. Perhaps markets are simply waiting to see the actual effect of the tightening, or perhaps even more tightening will be needed. Hopefully it is the former.
Central bank scope-creep reverses
Over the past several decades, central banks experienced considerable scope creep. Having primarily targeted inflation in the early 1990s, their focus shifted toward economic growth variables once inflation appeared to be structurally slumbering.
Financial stability matters were then added after the global financial crisis. More recently, subjects as diverse as inequality, climate change and cryptocurrencies began to play a more prominent role in central bank machinations. Some of these concerns have arguably contributed to today’s excessive inflation.
Of course, issues such as inequality and climate change are important – but mostly outside of the scope of monetary policy. Inequality concerns arguably contributed to the recent policy error to the extent that there was an attempt to push unemployment lower than normal since the most disadvantaged groups tend to be employed last.
Cryptocurrencies are plausibly relevant to monetary policy, but have taken resources away from more pressing matters in recent years.
After having been burned during the global financial crisis, central banks may have over-weighted financial stability variables since then – thus inadvertently underweighting inflation. In the context of the pandemic, that meant flooding markets with money so as to avoid a financial meltdown.
Lastly (and of greatest relevance to this analysis), sometimes monetary policy must disregard what is best for the economy in the short run if it is to do what is best for inflation. When those interests diverged around the middle of last year, central banks stuck to the economy maximizing strategy rather than pivot to the inflation-reducing approach of tighter monetary policy. That was a mistake.
Central banks are now aware of this error and are rapidly narrowing their focus to inflation and inflation alone. While one might expect some broadening again once this crisis is past, it seems unlikely that central banks will become quite so unfocused again.
-With contributions from Vivien Lee, Andrew Maleki and Aaron Ma