Business cycle signals recession?
Our U.S. business cycle received its quarterly update, revealing quite an interesting set of developments. The headline is that the scorecard has finally pivoted from its prior ‘end of cycle’ reading to ‘recession’ (see next chart).
U.S. business cycle shows weaker ‘end of cycle’ claim
As of 07/28/2023. Calculated via scorecard technique by RBC GAM. Source: RBC GAM
But the results require more discussion than that because the interpretation is not entirely straight-forward. Interestingly, the score for ‘end of cycle,’ which dominated over the prior several quarters, didn’t just recede but plummeted. Some of the points transferred to ‘recession,’ as you would expect, but an even larger share shifted backwards to ‘late cycle’ – meaning that while a significant number of economic variables continued to march forward, an even larger number moved backwards. The result is a split distribution in which ‘recession’ is the best single guess, but an alternate interpretation is that there are still several quarters before a recession.
In our view, the most logical interpretation is that a recession is now quite near at hand, albeit with a particularly large serving of imprecision given the scorecard’s mixed signals.
Furthermore, one can debate the meaning of the ‘recession’ signal. Does it mean that a recession is already underway? This seems unlikely based on our reading of current economic data. The ‘recession’ signal is most logically interpreted as meaning that a recession is on the cusp of beginning.
But a third, less orthodox, interpretation would be that this is as bad as it gets this cycle. This means that the cyclical trough wasn’t quite bad enough to induce an economic contraction. The fact that ‘start of cycle’ is now capturing a rising 8% of the vote provides tentative support for the view that a recovery is about to begin.
Despite this radically different possibility, in our view the most logical interpretation is that a recession is now quite near at hand, albeit with a particularly large serving of imprecision given the scorecard’s mixed signals.
Recession risk falls
Despite the business cycle’s ominous warning, economic news actually improved in recent months (see next chart). Combined with falling inflation, the consensus believes the recession risk has fallen significantly in some countries (see subsequent chart). Our own assessed recession risk for the U.S. over the next year has fallen from 80% a quarter ago to a more moderate 65% today.
Daily news sentiment improves
As of 07/23/2023. Sources: Federal Reserve Bank of San Francisco, Macrobond, RBC GAM
Probability of recession for some countries has come down
As of 07/28/2023. Median probability of recession based on latest forecasts submitted to surveys conducted by Bloomberg. Sources: Bloomberg, RBC GAM
Still, the actual economic data – in contrast to news about the economy – improved by much less than this. Many measures remain quite soft (see next chart).
U.S. business expectations remain lower
As of June 2023. Principal component analysis using National Federation of Independent Business (NFIB) optimism and business conditions outlook. Institute for Supply Management (ISM) Manufacturing and Services new orders, and The Conference Board CEO expectations for economy. Sources: The Conference Board, ISM, NFIB, Macrobond, RBC GAM
We have been reviewing a range of economic models and historical precedents for insight into the most important question of the day: whether the recession window is already closing or merely starting to open.
It was initially daunting (for those forecasting a recession, at least) to learn that the great majority of published central bank econometric models estimate that the effect of a rate hike is fully absorbed into the economy within a year. Taken at face value, that would mean the drag we’ve been experiencing from monetary policy is about to start easing as we approach the one-year anniversary of the most intensive period of rate hikes.
However, this is only true in a very narrow set of circumstances. When these models experience a sudden interest rate shock, they presume that the shock will abate quickly, meaning that the policy rate rapidly returns to its baseline state. In practice, that doesn’t usually happen, and it hasn’t happened this time: not only have policy rates failed to reverse, but they continue to push higher.
Our own historical analysis of past tightening cycles finds that the timespan from the first rate hike to recession averages 27 months: this calls for a mid-2024 recession.
With more realistic assumptions, such models – including the one we use for our own modelling – estimate that each rate hike delivers a fairly steady headwind lasting about two and a half years. If we stack the effect of all the rate increases we’ve had on top of one another, the monetary drag should be incrementally greater over the coming year than it was over the past few quarters. The recession window is still actively opening wider.
To be sure, an incrementally greater drag hardly guarantees a recession. The economy has so far defied recession despite the current level of drag. This is why the recession odds are well short of 100%.
But a large literature review published in the International Journal of Central Banking found that the average lag from rate hikes to a negative effect on inflation is a whopping two to four years. According to this evidence, any recession would be ahead of schedule for quite some time.
Similarly, our own historical analysis of past tightening cycles finds that the timespan from the first rate hike to recession averages 27 months (see next table): this calls for a mid-2024 recession.
Finally, just from a common-sense perspective, it is patently obvious that the full effect of rate hikes has not yet been felt within the economy given that many mortgage-holders and other borrowers have not yet rolled into higher interest rates.
Historically, recessions come with a long lag after tightening
U.S. business cycles and tightening cycles. Sources: Federal Reserve, National Bureau of Economic Research, Macrobond, RBC GAM
Frustratingly, in economics, there isn’t much precision with anything. It is an inexact science. It is sorely tempting to abandon a recession call when the economy keeps refusing to quit, inflation is falling and rate hikes are plausibly nearing an end. But the magnitude of tightening already delivered is theoretically sufficient to create a recession and the lags involved still leave plenty of room for a recession.
Similarly, our aforementioned business-cycle work and our collection of recession heuristics (see next chart for that latest version) all continue to argue that a recession is still more likely than not. We must not let emotions obscure such inputs.
Recession signals point mostly to ‘yes’ or ‘likely’: we estimate 65% chance over the next year
As of 07/24/2023. Analysis for U.S. economy. Source: RBC GAM
New growth forecasts upgraded
In the prior MacroMemo we conveyed a delay in the timing of the anticipated recession from the second half of 2023 to the final quarter of 2023 and the first quarter of 2024.
We have further updated our growth forecasts over the intervening weeks, with the result that the recession depth is also now assumed to be slightly shallower than before. For the U.S., the peak-to-trough decline recedes from -1.4% to -1.1%. It is a similar adjustment in other developed countries. In qualitative terms, this takes the anticipated recession from a ‘mild to middling’ decline to more of a ‘mild’ decline.
On the whole, the delayed recession pushes 2023 growth forecasts higher and 2024 growth forecasts lower.
We remain below consensus in our growth forecasts, and – in part due to the softer growth forecast – we continue to forecast that inflation falls more quickly than the market assumes, much as it has in recent quarters.
Notwithstanding the diminished recession intensity, a tally of the new economic forces vying for influence is fairly evenly split between positives and negatives.
Positive developments include:
- The economy has continued to grow, defying recession for another quarter. This is a mathematical positive for GDP (Gross Domestic Product) growth in 2023, and also a signal that the economy retains a measure of resilience.
- Inflation fell quite nicely, reducing a corrosive effect against growth.
- Housing remains in a rejuvenated state.
- Excitement about generative AI may not boost productivity growth in the short run, but it can moderately improve R&D and capital expenditure spending.
- The stock market has been rising, generating a small positive wealth effect.
Conversely, negative developments include:
- Central banks delivered more rate hikes than previously envisioned.
- There is still a significant lagged effect from prior rate increases.
- The U.S. debt ceiling deal brought a modest amount of fiscal drag to that country.
- S. student loan payments will recommence in August, costing tens of millions of Americans hundreds of dollars per month.
- The Chinese economy continues to sputter, underwhelming expectations.
- There was another pinch of banking stress over the past quarter, albeit way back in May.
We remain below consensus in our growth forecasts, and – in part due to the softer growth forecast – we continue to forecast that inflation falls more quickly than the market assumes, much as it has in recent quarters (see next chart).
RBC GAM forecasts vs. consensus for 2024
Deviation measured as difference between RBC GAM forecast (07/24/2023) and consensus forecast (July 2023). Sources: Consensus Economics, RBC GAM
Inflation still falling
Inflation continues to fall quite nicely. The U.S. printed a 3.0% year-over-year (YoY) CPI (Consumer Price Index) number several weeks ago. Now Canada has followed suit with a 2.8% YoY reading for June. Eurozone inflation is lagging, but down to 5.3% YoY. Small businesses have become somewhat less anxious about inflation (see next chart).
Inflation remains biggest small business problem
As of June 2023. Shaded area represents recession. Source: NFIB Small Business Economic Survey, Macrobond, RBC GAM
Improvement likely to slow
But we warn that the rate of inflation improvement from here should slow. The price of West Texas oil has surged from less than US$70 per barrel to nearly US$82 over the past month, providing an upward pressure via gasoline prices.
The base effects also become less favourable. Inflation started to slow precisely a year ago, so the annual print won’t be discarding giant monthly gains from 12 months earlier anymore.
Business pricing plans have also firmed up slightly (see next chart).
Fraction of U.S. businesses planning to raise prices increasing again
As of June 2023. Shaded area represents recession. Sources: NFIB Small Business Economic Survey, Macrobond, RBC GAM
Consistent with all of this, one of our key real-time inflation sources suggests that inflation may not have improved any further in July (see next chart).
U.S. Daily PriceStats Inflation Index stalled
PriceStats Inflation Index as of 07/22/2023, CPI as of June 2023. Sources: State Street Global Markets Research, RBC GAM
Food inflation risk
There are also upside risks to inflation ahead. Prominently, although food inflation is slowing (see next chart), there is a risk this happy trend stalls out. Russia is no longer permitting Ukrainian grain to transit via the Black Sea, increasing global grain prices moderately (see subsequent chart). The issue here is less the price increase so far, and more the risk that supply problems get even worse. Also, the embargo has foodstuff implications that extend beyond wheat. A further complication is the recent global heat wave, which could undermine harvests across at least three continents.
U.S. food inflation has declined
As of June 2023. Shaded area represents recession. Sources: U.S. Bureau of Labor Statistics, Macrobond, RBC GAM
Wheat prices jumped as Russia pulled back out of Black Sea grain deal
As of 07/27/2023. Sources: S&P Global, Federal Reserve Bank of St. Louis, Bank of Canada, Macrobond Financial AB, Macrobond, RBC GAM
Wage-price spirals remain another key upside inflation risk. Although the official measure of U.S. hourly wage growth is down from a high of 5.9% to 4.3% today, this exaggerates both the improvement and the relative tameness of today’s wages. Composition effects distort the numbers as lower skilled workers first left and then re-entered the workforce. This made average wages first appear higher and now lower.
The Atlanta Fed’s wage growth tracker controls for these issues. It finds that the actual rate of wage growth remains considerably faster than the official measure, and has slowed more modestly, from a peak of 7.1% to 6.0% today (see next chart). Our measure of wage expectations argues the rate of wage growth should slow significantly from here, though actual wages have lately been ignoring its prophecies.
Wage pressure easing in U.S.
Atlanta Fed Wage Growth Tracker as of June 2023, wage expectations as of July 2023. Wage Pressure Composite constructed using business intentions to raise wages. Shaded area represents recession. Sources: Macrobond, RBC GAM
As we have discussed in the past, U.K. wage growth continues to accelerate in a worrying way (see next chart). This helps to explain why the Bank of England is the one major developed-world central bank with significant monetary tightening still expected.
Growth of U.K. average weekly pay accelerates
As of May 2023. Sources: U.K. Office of National Statistics, Macrobond, RBC GAM
After such high inflation and with extremely low unemployment, it is a time of mounting labour discontent in many markets.
- A recent U.S. Teamster Union strike was barely avoided.
- There are widespread expectations that workers will strike on September 14 against one of the major U.S. automakers.
- The number of U.S. work stoppages is beginning to grow (see next chart).
- The U.K. has suffered a truly startling number of strikes over the past year (see subsequent chart).
Number of work stoppages in U.S. has gone up since 2020
As of June 2023. U.S. work stoppage data includes work stoppages involving 1,000 or more workers. Sources: Bureau of Labor Statistics Macrobond, RBC GAM
Number of work stoppages in U.K. has skyrocketed since before the pandemic
As of May 2023. Stoppages data includes those caused by labour disputes between employers and workers, or between workers and other workers, connected with terms and conditions of employment. Sources: U.K. Office for National Statistics (ONS), Macrobond, RBC GAM
Inflation spikes tend to be worse in U.K.
Given that inflation in the U.K. has recently been worse than for other developed countries, we investigated prior bouts of high inflation to see whether this was just bad luck or part of a broader pattern.
The tentative conclusion is that the U.K. does appear to be more structurally vulnerable to high inflation (see next chart). Not only did British inflation reach a higher peak in 2022, but it also peaked well above the U.S. in the early 1990s and during most of the inflation spikes in the 1970s and early 1980s.
U.K. inflation spikes historically worse than U.S.
As of May 2023. Sources: Bureau of Labor Statistics, U.K. Office of National Statistics, Macrobond, RBC GAM
In fairness, the U.K. inflation rate was relatively unremarkable across the remainder of the 1990s, the 2000s and most of the 2010s, meaning that this isn’t an observation that has application in every cycle.
The best all-weather explanation is that the U.K. is reliant on volatile imports as a small, open, island nation. Brexit provides a further explanation for the latest spike, while strong union membership helps to explain the inflation spikes of the 1970s. At that time, the U.K. unionization rate was above 50%. This is a significantly diminished force today, though clearly still relevant. We would argue that, absent special forces, global inflation spikes in the future should be only slightly worse for the U.K. than the U.S. But, first, let’s deal with the current one.
Central banks near the finish line?
After delivering an additional burst of monetary tightening in recent months that exceeded earlier expectations, many developed-world central banks could finally be nearing the finish line.
The U.S. Federal Reserve delivered the anticipated 25bps rate increase last week, raising the upper end of the fed funds rate target range to 5.50%. Although the accompanying communiqué upgraded the assessment of the economy, the trailing press conference was more dovish. Comments related to the restrictiveness of monetary policy, the long distance that monetary policy has already travelled, emphasizing a data dependent approach for the future that far from assures additional monetary easing.
When central banks have historically engineered a tightening cycle, paused and then recommenced tightening (as the Fed recently did), there is more than a single rate hike 80% of the time.
Financial markets believe the Federal Reserve is now likely done, with just one-fifth of another 25bps rate increase priced for the autumn. That is a credible view, especially in light of cooperating inflation. However, let us not forget that the Fed dot plots from June had pointed to a further 25 basis point rate increase before the year is complete.
Moreover, when central banks have historically engineered a tightening cycle, paused and then recommenced tightening (as the Fed recently did), there is more than a single rate hike 80% of the time. Another rate hike is not off the table, especially if the economy were to remain strong and inflation ceases to improve as materially in the coming months.
Elsewhere, the Bank of Canada’s minutes revealed a highly data-dependent approach that arguably tilts toward no action at the next meeting: “They agreed they were prepared to raise the policy rate further if inflationary pressures did not ease as projected … But they did not want to do more than they had to.” (BOC, 7/26/2023)
The European Central Bank raised its deposit facility rate by 25bps to 3.75%, and also took a highly data dependent stance.
The Bank of England constitutes an exception, with the market still pricing three-plus 25bps rate increases from the current 5.00% setting.
The Bank of Japan, meanwhile, continues to march to a very different drummer. It never experienced the initial inflation surge felt by other countries, only recently feeling some inflation heat. Furthermore, after decades of undershooting inflation targets, the country is much more tolerant than most of higher inflation as a tool to revive long-term inflation expectations.
But its indulgence of inflation is not infinite. Japan continues to print money and distortions abound from a long history of negative interest rates. The Bank of Japan first relaxed its system of yield curve controls in December, allowing the 10-year bond yield to reach as much as 0.50%, higher than the prior 0.25% limit. Last week, the Bank of Japan effectively further increased the limit, indicating it will offer to buy 10-year Japanese government bonds at up to 1.0% in fixed-rate operations. Confusingly, the 0.50% rate remains the ‘reference’ level, but it is no longer a ‘rigid limit.’ The result has been that the Japanese 10-year yield now trades above the prior 0.50% limit, at the highest rates in nine years (see next chart).
Bank of Japan cautiously tweaks yield-curve control policy as global rates rise
As of 07/28/2023. Sources: Bloomberg, RBC GAM
We assume Japanese yields will continue to edge higher over time as Japan makes further policy tweaks.
Looking further ahead – and back to non-Japanese developed countries – it is notable that some emerging market economies are starting to cut their policy rates. Chile just delivered a 100 basis point rate cut, and Brazil is expected to initiate its own easing cycle in August. Others will likely follow.
This is potentially relevant to the developed-world because emerging market central banks are often the canary in the mineshaft. They were raising rates well before developed-world central banks in 2021 given a greater sensitivity to food prices and capital flows. They may now be in the position of signaling the beginning of an easing cycle. In fairness, developed nations are likely still many months away from any pivot.
A fiscal thought
We have noted before that fiscal deficits in a wide range of countries remain uncomfortably large for this late stage of the economic cycle (see next chart). The U.S. had begun turning in a constructive direction (and will benefit shortly from debt ceiling-related restrictions). In the meantime, its deficit has again shot wider (see subsequent chart).
Significant structural fiscal deficits persist
International Monetary Fund (IMF) projections for year 2023. Sources: IMF World Economic Outlook, April 2023, Macrobond, RBC GAM
U.S. fiscal deficit has expanded back out
As of June 2023. Sources: Macrobond, RBC GAM
Governments are under a great deal of pressure to spend for many reasons:
- Service existing debt.
- Pay for more muscular militaries.
- Support aging populations.
- Finance green initiatives.
- Engage in newly popular industrial policies.
At the same time, the bond market is becoming more sensitive as central banks unwind the quantitative easing operations that long muted market forces.
There are a number of ways to resolve large deficits and heavy debt loads.
The most optimistic strategy is simply to grow one’s way out of the problem. But it is easier said than done to accelerate GDP growth. In the present situation, the short-run growth risk extends more toward a weaker economy if a recession appears.
Of course, it is plausible that new technologies will likely accelerate productivity over the long run. But it isn’t certain, and the magnitude isn’t yet clear. Furthermore, one would need a truly heroic and sustained acceleration in GDP growth to deal with deficits currently in the 4-7% of GDP range.
The standard deficit-reduction strategy is simply to reduce the imbalance between government revenue and expenditures. This is never easy given that tax hikes and spending cuts are both highly unpopular with voters. The gentle version is simply to limit the growth of spending more judiciously than normal. This will have to be part of any solution.
Any solution will have to come primarily from fiscal austerity, but with a bit of help from faster long-term growth, interest rate repression and extra inflation.
The other solutions are interest rate repression and tolerating a bit more inflation. We have repeatedly argued that this is likely to remain an era of fairly low interest rates, in significant part because of how much debt there is. High debt doesn’t result in low interest rates by itself. Rather, it creates an incentive to run lower rates than otherwise to keep the debt-servicing burden from becoming too onerous.
Similarly, we believe long-run inflation could be inclined to run slightly above target, primarily due to factors having nothing to do with debt. De-globalization and climate change top the list. However, there could be a bit more tolerance for this extra inflation to the extent it also helps to reduce the burden of high public debt. It does so, incidentally, in two ways:
- It boosts the nominal GDP growth rate (this helps the debt-to-GDP ratio).
- It effectively reduces the real interest rate for already-locked-in government bonds.
The bottom line is that any solution may have to come primarily from fiscal austerity, but with a bit of help from faster long-term growth, interest rate repression and extra inflation. So far, most countries have not been willing to pursue that first option, though their attitude will likely change once bond markets become more concerned. Countries with higher deficits will be forced to pay larger risk premiums on their debt, and worse fates are possible.
Constructive political signals from China
China’s politburo meeting in July failed to unveil any major new fiscal stimulus to revive the Chinese economy. However, it did arguably lay the groundwork for more support later. At a minimum, the government is more concerned than it was a quarter ago: economic problems were acknowledged – in particular, insufficient domestic demand and corporate challenges.
In typically cryptic fashion, the politburo pledged to strengthen countercyclical adjustments – code for economic stimulus of some form. Commitments included to strengthen flagging domestic demand, mentioning both the auto sector and electronics, among others.
The housing market may also receive some support given that the prior line of ‘housing is for living in, not for speculation’ was omitted.
Finally, the government committed to resolving the local government debt problems that we wrote about in our 27 June MacroMemo. This is a big deal, as there may be US$10 trillion or more of Chinese local government debt that is struggling given the weaker housing market. Of course, government money that remedies local government debt is money not doing other productive things in the economy.
India’s future promise
India is an attractive country from a long-term investment and economic perspective. Amazingly, the International Monetary Fund (IMF) forecasts that India will be the second-largest driver of global economic growth through 2028, ahead of the U.S. (see next chart).
China to remain the top driver of world growth
Based on International Monetary Fund (IMF) forecast from 2023 to 2028. Sources: IMF World Economic Outlook, April 2023, Macrobond, RBC GAM
From a growth rate perspective, the IMF additionally forecasts that India will significantly outgrow China over this coming period (see next chart), making it the fastest growing large economy in the world. Just to clarify, China’s economy is so much bigger than India’s that its relatively more paltry percent growth rate will still yield a larger dollar gain in output each year. But India will be getting ever closer as a percent of China’s output, and its immense population gives it great promise.
Economic growth: India likely to grow faster than China
Sources: IMF World Economic Outlook April 2023, Macrobond, RBC GAM
To be sure, there is much that India must still work on. Its infrastructure lags not just China but many of its developing-nation peers (see next chart). Furthermore, the country has a fairly high level of public debt for a developing nation (see subsequent chart).
Ranking of infrastructure by country shows India lags China and other developing nations
As of 2022. Sources: Global Innovation Index, Macrobond, RBC GAM
India government debt remains high
As of Q4 2022. Sources: The Bank for International Settlements, Macrobond, RBC GAM
But as international companies pursue ‘China Plus One’ policies, India is a natural recipient of a healthy share of that diversified manufacturing. Those initial forays will help to build out the infrastructure, potentially setting in motion a virtuous circle of expansion. The Indian government has laid the foundation for a period of good growth in several regards, including a new system of electronic transactions that has dragged commerce into the 21st century (see next chart).
India Unified Payments Interface value rises
As of May 2023. Sources: Reserve Bank of India, Macrobond, RBC GAM
B.C. port strike – take two
We wrote two weeks ago that the B.C. port strike had been resolved via collective bargaining. We were right at that particular moment, but then wrong as the union leadership later rejected the deal, then right as the union leadership backtracked, then wrong as union membership rejected it in a vote, and then right again as a modified deal was struck.
At the rate things are going, we may yet be wrong and then right several more times. This new tentative deal needs to be voted on. The federal government could become more involved given the strong national interest if there is a ‘no’ vote. And in an extreme scenario, the courts could even become involved given recent Canadian court rulings that limit the use of back-to-work legislation. Hopefully it doesn’t come to that, and the best guess at present is that the new deal will stick.
The implications remain largely the same as our last assessment: we figure the economy could lose several tenths of a percentage point of output, with that loss not fully recovered for several months. But the risk is clear that the damage could be greater, either because the latest deal isn’t ratified, or because of mounting evidence that the initial damage may be more lasting than initially imagined. Canada must grapple with supply chain problems once again, if only temporarily.
With contributions from Vivien Lee, Thao Le and Aaron Ma