July 2022
Summary
The classic trade-off between economic growth and inflation is re-emerging for the Fed
Although forward looking indicators point to slowing growth, there is an unusual amount of uncertainty in this cycle around the sensitivity of growth and inflation to higher rates
The recessions of the 1960s, 1970s and 1980s may offer better clues as to how sensitive inflation will be to slowing growth than the economic downturns in the last two decades
The current political support for higher rates will likely diminish as the economy slows in response to the higher rates
As the cost of reducing inflation becomes apparent, expect to hear more calls for the Fed to tolerate a higher rate of inflation
Investment Implications
The Fed’s current motto of being “humble and nimble” means a likelihood of greater volatility in interest rate and fixed income markets
Greater macro and interest rate volatility should offer opportunities for tactical and directional trading strategies
Markets appear to be under-pricing the risk that there may be a structural drift higher in the Fed’s inflation target; financial repression is likely to continue to be a theme
Financial markets continue to struggle to adjust to the dynamics and implications of a new type of tightening cycle.
In January, in Is this the end of gradualism? we wrote that having delayed withdrawing stimulus the Fed would have to abandon the gradualist approach to monetary policy that characterised each tightening cycle this century.
After raising rates 25bps, 50bps and 75bps at consecutive meetings since March, the fastest pace of tightening since the 1980s, there is little doubt that we are into a new era of central bank policy making.
Previous Fed tightening cycles since 1988
Source: FRED, Archive Capital
The key differentiator in this tightening cycle is that the Fed is using monetary policy to slow the economy to bring inflation down. At the start of every other tightening cycle since 1994, core PCE was 3% or less at the start of the cycle and policy was aimed at ensuring the economy didn’t overheat rather than actively attempting to reduce inflation.
Core PCE (%, yoy) since 1990
Source: FRED, Archive Capital
What this means is the starting point is one of deeply negative real interest rates and the last two decades may offer little precedent for this cycle.
Here we assess three key questions that will be critical in this cycle:
1. How sensitive will growth be to monetary tightening?
2. How sensitive will inflation be to weaker economic growth?
3. What is the ultimate target for inflation?
How sensitive will growth be to higher rates?
At any point in time in financial markets there is a spread of bulls and bears, doves and hawks and optimists and pessimists. That spread of opinion feels unusually large at the moment.
Take crude oil as an example: JPMorgan sees potential for crude oil to rise to $380, while Citi are forecasting a decline to $65 by year end.
That uncertainty is also evident in relation to the broader economic outlook. Growth pessimists see growth rather than inflation as the primary challenge with some forecasting rate cuts by the Fed by year end. Others see inflation becoming more entrenched with the risk that rates may need to rise as high as 8% to bring inflation back in line.
The consensus is that the economy will slow but by how much, in what time frame and for how long.
As Chair Powell pointed out at his recent congressional testimony, ultimately the Fed wants to bring growth down to below trend but not below zero. But the question from a market perspective is how high will interest rates need to go and will the Fed be successful in achieving the soft landing.
In the aftermath of the 8.6% print for May CPI, interest rate futures moved to price in a much more aggressive policy path from the Fed. Dec 2022 Eurodollar futures declined by over 100bps between 26 May and 14 June with about 60bps coming in the two days on and after the CPI release. The Fed validated the expectation of a more aggressive tightening cycle in June with a 75bps hike, the first of this magnitude since 1994.
However, since the June FOMC meeting, investor focus has shifted to concerns about growth rather than inflation. US 10-year yields are currently about 60bps off their highs and Dec 2023-Dec2022 Eurodollar future spread point to about 75bps of easing in US short rates in 2023 (a shift from +10bps of tightening in 2023 in early June).
December 2022 - December 2023 Eurodollar futures spread
Source: CME, Archive Capital
The more aggressive approach by the Fed, weakness in equities, the sharp falloff in consumer confidence, the decline in growth sensitive commodities such as copper, signs of a slowdown in the recent manufacturing ISM new orders index and the sharp increase in mortgage rates are all forward looking indicators pointing to a slowdown.
US Manufacturing ISM and Copper prices
Source: FRED/Archive Capital
Currently, Eurodollar futures markets are pricing a rise in rates to just over 3.5% by year-end but for the Fed to be cutting rates in 2023.
Implied US 3-month rate (%) from Eurodollar futures
Source: CME, Archive Capital
Does that look reasonable?
A rise in rates to 3.5% would mark a more aggressive hiking cycle than between 2015-2018 but a smaller net increase in rates than in 2004-2006 even though inflation is substantially higher now than in that period.
In 2019 the Fed lowered rates seven months after its last rate increase the 2015-2018 cycle: that may explain the expectation of a quick unwind of some of the tightening in 2023.
Three factors make it particularly difficult to assess how much growth might slow in response to higher rates:
1. Rate increases have been unusually quick but from exceptionally low levels. It’s the first time since 1994 the Fed has raised rates by 75bps. Indeed, having initially being slow to withdraw stimulus, the current pace of rate increases is faster and more aggressive than any rate hiking cycle since the 1980s.
Cumulative increase in Fed funds rate in each Fed tightening cycle since 1988
Source: FRED, Archive Capital
2. Higher interest rates (and the start of quantitative tightening) have translated into higher 10-year yields and a broader tightening of financial conditions. In the 2004-2006 tightening cycle, the Fed funds rate rose by 4.5%. However, in that cycle bond yields initially did not rise with interest rates (Greenspan’s bond market conundrum) and consequently financial conditions didn’t tighten until the latter stages of the tightening cycle. This time around US-10 year yields have risen close to 3% from the lows in March 2020 the largest increase since the late 1980s.
Increase in US 10-year yield from previous low
Source: TradeView/Archive Capital
Weakness in equities, bonds and credit have tightened financial conditions and reduced wealth. At the same time consumers’ real incomes have been eroded by higher energy prices. Historically such rapid increases in yields have been associated with notable declines in the PMI so a slowdown does look likely. That said, yields rose a similar amount in 1994 and the Fed did achieve a soft landing in that case.
US 10-year yields and Manufacturing ISM
Source: FRED, Archive Capital
3. However, what is unusual about this cycle is real interest rates (as measured by current inflation) are still negative. Even taking core PCE rather than headline CPI as the relevant measure of inflation and interest rates are still negative in real terms. In standard economic textbooks it is real interest rates that typically drive behaviour.
Historically, negative real rates have tended to occur at the end of recessions rather than precipitating recession (such as 1971,1975 and 1980). Negative real interest rates could motivate consumers to borrow to bring forward consumption even with rising interest rates. It is noteworthy that consumer credit growth has been rising recently (although possibly due to the squeeze on real incomes).
In summary, although many indicators now point to a slowdown, the economy is particularly strong at the moment and there is an underlying support from real interest rates so the timing and magnitude of the downturn remains open for debate.
How quickly and how much growth might slow in response to higher rates is only the first part of the current investment puzzle. Understanding how much inflation may slow in response to weaker growth will be key for understanding the Fed’s reaction function.
In the Global Financial Crisis and during the initial phase of COIVID CPI fell sharply as unemployment rose and the economy weakened. However, in this cycle because (1) wages may respond with a lag to recent inflation and (2) supply side challenges may remain there is a risk that inflation remains sticky even as growth weakness.
The inflation-unemployment trade-off
The relationship between inflation and unemployment, as captured by the Philips Curve, has long been a controversial and hotly debated topic in economics.
In the 1960s, Philips’ observation that there appeared to be a relationship between wage growth and unemployment, and then more generally between overall inflation and unemployment, encouraged a belief amongst policy makers that the relationship could be exploited i.e. by accepting a higher level of inflation unemployment could be maintained at a permanently lower level. In effect the question was how much inflation should be tolerated to bring unemployment down.
US Philips Curve (CPI versus Unemployment rate)
Source: FRED, Archive Capital
However, the 1970s proved that the relationship was not stable. Because workers might adjust to repeated monetary easing with higher inflation expectations and wage demands the result would be accelerating inflation without lower unemployment. This spawned the concept of the NAIRU (the non-accelerating inflation rate of unemployment). The stagflation of the 1970s appeared to support this interpretation.
However, over time as central bankers shifted their focus to price stability and influencing inflation expectations, there was evidence that the Phillips curve was still in existence but due to various factors the curve had flattened. Whereas Hooper et al (2019) estimated that the Phillips curve had a -0.7 slope for the 20 years to the early 1990s, by 2019 the estimate was about -0.2.
This interpretation was part of the rationale for the Fed running the economy hot in 2019 and post COVID, as there was a general belief that a tight labour market would not prompt higher inflation.
However, post-COVID we have again seen a shift in the relationship between unemployment and inflation. Supply side constraints coupled with strong demand fuelled by easy money and stimulative fiscal policy has pushed up inflation substantially as unemployment has fallen.
One explanation for why inflation may have taken off is that at very low levels of unemployment the curve may be non linear and inflation will rise quickly in a tight labour market. Based solely on the unemployment rate, the labour market doesn’t look unusually strong versus 2019 but other labour market indicators like quits and job openings have suggested that it is unusually tight.
Why does all this matter to the current debate?
Economists and policymakers have long focused on the Phillips curve to answer the question how low can unemployment go without causing significant inflation, but in the current market the relevant question becomes how much must unemployment rise to get inflation back to target?
If inflation proves to be sticky then the cost of reducing inflation (in terms of lost output and rising unemployment may be substantial).
Indeed, looking at past periods when the Fed was actively trying to reduce inflation, we can see that inflation proved to be quite stubborn in the face of rising unemployment.
CPI, Core CPI and Unemployment during the US recessions (NBER definition)
between 1969 and 1991
Source: FRED/Archive Capital
For example in the 1970 recession, unemployment rose by 2.4% and edged slightly higher just after the recession. But CPI fell just -0.3% in the period and core CPI actually rose. A 2.5% increase in unemployment in the current US labour market would be a loss of about 4m jobs. The cost of the 1973-1974 recession was even greater, largely due to the oil crisis.
With inflation running at rates in excess of 8% and the economy booming there was widespread political support for higher rates. However, as the economy slows in response to higher rates the Fed will face the trade-off between unemployment and inflation.
The standard argument for why reducing inflation is more favourable than keeping unemployment low is that higher inflation impacts everybody in the economy whereas lower unemployment only benefits those people who are employed and would otherwise be unemployed if the economy slowed.
However that’s a somewhat superficial analysis. There are a number of winners and losers to higher inflation. Higher inflation favours debtors overs creditors. Higher inflation favours home owners over those searching for a home. Higher unemployment tends to disproportionately impact the lower classes. Reducing unemployment for marginalised groups had been a key objective for the Powell Fed in the pre-covid economy (remember at one point even a suggestion of targeting the rate of black unemployment).
Bear in mind that after his tenure at the Fed former Chair Arthur Burns attributed the inflation of the 1970s to a lack of political support for quelling inflation during that period; the cost of reducing inflation in terms of lost employment was deemed politically unpalatable until Volcker.
To the extent that reducing inflation involves a prolonged increase in unemployment coupled with lower equities (already the case) and possibly lower house prices it is easy to see how the political support for higher rates may diminish over time.
That’s why the Philips curve has become important again. A 2% rise in the unemployment rates to reduce inflation by 2-3% may be an attractive trade-off but a 3-4% rise in unemployment to achieve the same goal may prompt calls for a reassessment of the policy mix.
What is really the inflation target? 2% or 4%?
That brings us to the third significant uncertainty at the moment: does the Fed really want to get inflation back to 2%?
Publicly, the Fed continues to state that its target is to bring inflation back to 2%. However, we have already seen calls from economists for considering a relaxation of the 2% target.
In a recent column, “How low must inflation go”, Paul Krugman outlined how 2% became the standard inflation rate for price stability (largely because that was what the early inflation targetters such as the Reserve Bank of New Zealand chose rather than any economic rationale) and outlined a case for a higher inflation target.
The suggestion that central banks should consider targeting a higher level of inflation came to the fore after the Global Financial Crisis as central banks struggled to push real interest into negative territory without dealing with the negative consequences of negative nominal rates.
In 2010, IMF Chief Economist Oliver Blanchard outlined a case for a 4% inflation target. At the time it was an academic debate as regardless of what you thought was the correct inflation target, the big problem facing central banks was the inability to generate sufficient demand to push up inflation.
That’s no longer the case. One side effect of the current inflationary surge is that it offers the option to reset the definition of price stability or the inflation target to give the flexibility of having more negative interest rates when the economy next needs monetary stimulus.
As Krugman points out, from an economic perspective there is no real reason why the Fed or any other central bank should want to get back to 2%. 3%, 4% or 5% could all be just fine. The main argument for 2% is to maintain credibility.
How quickly central banks can reduce inflation is partially out of their hands (commodity prices and the supply side impacts of COVID will have a significant impact). If the perception increases that the cost of each incremental reduction in inflation is too large (in terms of lost output and unemployment) its easy to engage a scenario where the Fed may be willing to accept inflation in a 3-4% range.
5-year 5-year forward Inflation Expectation Rate (%)
Source: FRED/Archive Capital
From that perspective it surprising that 5year 5 year forward breakeven rates have fallen of late and remain broadly in the range of the last 20 years. The risk that the Fed accepts a permanent drift higher in inflation over time appears to be under-priced.
Summary
The current challenge facing the Fed of actively slowing the economy to bring inflation down is different to the policy challenges on the 2000s and 2010s and more akin of the policy challenge in the 1960s, 1970s and 1980s.
With a large degree of uncertainty around key macroeconomic relationships and a Fed focused on being “humble and nimble” there is an unusually high importance on the near term data on growth and inflation rather than projections 2-3 years out.
The Fed can no longer afford to passively respond even if their models suggest that inflation may moderate. That suggests the potential for an overreaction to near term data releases and greater volatility in interest rate markets in particular but more broadly across asset classes.
Such an environment may continue to be favourable macro orientated strategies and directional strategies like trendfollowing.
The political support for economic pain to reduce inflation may diminish over time and there is a risk of a structural shift higher in the Fed’s inflation tolerance which seems under-priced at the moment.
The information and commentary presented in this website by Archive Capital is of a general nature for information and education purposes. It is not to be used or considered as a recommendation to buy, hold or sell any securities or other financial instruments; and does not constitute an investment recommendation or investment advice.
This material is: (i) for the private information of the reader, and Archive Capital is not soliciting any action based upon it; (ii) not to be construed as an offer or a solicitation of an offer to buy or sell any security in any jurisdiction where such an offer or solicitation would be illegal; and (iii) based upon information that Archive Capital considers to be reliable.
The information in this website has been obtained from sources believed to be reliable, but its accuracy and completeness have not been verified and are not guaranteed. The opinions, estimates and projections constitute the judgment of Archive Capital and are subject to change without notice.
Archive Capital does not warrant or represent that the information is accurate, complete, reliable, fit for any particular purpose or merchantable; and does not accept liability for any act (or decision not to act) resulting from the use of this information and related data.
Comments