Recession delayed is recession denied: fundamentals of the most hated US equity rally in history
Plus, Meta’s Threads is a win for federalism
Half-way through the year and the run of US data continues to hold-up remarkably well to the Fed’s tightening cycle, consistent with our contrarian take on the turn of the year consensus that favoured a US recession in 2023.
The no near-term recession case is unlikely to be significantly challenged by what we expect will be a mixed US employment report on Friday. The market is expecting a 200k gain in non-farm payrolls, with the unemployment rate seen falling from 3.7% to 3.6%. We are more upbeat on payrolls, with a forecast of 239k, but more downbeat on the unemployment rate, expected to remain steady at 3.7%. The pace of employment growth continues to slow, but the monthly net gain in employment remains robust by pre-pandemic standards.
Don’t under-estimate labour supply
A key underpinning of the recession consensus is the mistaken assumption that the unemployment rate must rise to tame inflation. But inflation has already moderated substantially while the unemployment rate holds near its lowest levels since 1969. As Adam Ozimek notes in a piece in The Atlantic, the expectation that the unemployment rate must rise to lower inflation ignores the supply of potential workers that can be tapped to meet the excess demand for labour. This is just a special case of the broader argument for expanding the supply-side of the economy to resolve an excess demand problem, rather than crushing the demand-side with monetary policy.
As Ozimek notes, policymakers have consistently erred in their assumptions about the scope to draw in workers on the margins of labour market participation:
After the Great Recession, many economists believed that the share of people who simply wouldn’t work had permanently increased. The supposed reasons were as varied as health problems and the high quality of video games. And yet, as the recovery progressed, more and more people returned to work, disproving the labor-market pessimists year after year. Wages rose at a pace that delivered real improvements in standards of living, but without inflationary pressure.
One of the reasons Australian policymakers kept monetary policy too tight pre-pandemic was a view that the then unemployment rate of around 5% was so low that it must eventually give rise to inflation pressures via faster wage growth. Needless to say, those pressures never eventuated, at least not pre-pandemic. Not only were the assumed price-wage dynamics wrong, but there were upside surprises in labour force participation. By September 2019, Australia’s labour force participation rate had risen to a then record 66.2%. In May this year, it made new record highs at 66.9%, continuing the post-pandemic labour market recovery.
Is US monetary policy even tight?
Another assumption underpinning the near-term recession case is that US monetary policy has been excessively tightened. This has prompted quite a bit of head-scratching and not just in the US. This FT piece on Why are interest rate rises not taming inflation? is representative. An obvious answer to that question is the one given by Scott Sumner: interest rates don’t tell you much about the stance of monetary policy. As Scott argues:
There’s no mystery here—easy money is generating fast NGDP growth, and that’s why core inflation remains stubbornly elevated.
This is just flip-side of the pre-pandemic assumption that monetary policy was easy because interest rates were low. The New York Fed’s Global Supply Chain Pressure Index is now well below its historical average, so we can no longer rationalise inflation as a largely supply-side problem:
Sumner’s market monetarist view is at odds with those monetarists like Scott Grannis who focus on monetary aggregates, but Sumner would argue that monetary aggregates by themselves are not a reliable guide to the stance of monetary policy. The obvious counter to Sumner’s view that is also consistent with his ‘price of money’ approach is the ongoing yield inversion (which can be interpreted as a relative price). But for all the focus on the yield curve, there are other recession indicators with a good track record that imply a very low probability of recession.
Jeremy Pidger’s dynamic-factor Markov-switching model based on four coincident variables suggests the probability that the U.S. economy was in recession in April 2023 was just 0.6%. Historically, three consecutive months of smoothed probabilities above 80% has been a reliable signal of the start of a new recession, while three consecutive months of smoothed probabilities below 20% has been a reliable signal of the start of a new expansion:
In contrast to standard term spread-derived probability models, this model’s reliance on coincident indicators might be thought to be somewhat tautological in estimating a low probability of recession, but nonetheless highlights the disconnect between recession sentiment and recent data.
Because of its high salience for consumers and firms, high inflation is very bad for sentiment, which helps explain why survey-based measures of sentiment and activity have been underperforming the hard data like employment.
Fundamentals of the most hated equity rally in history
The YTD rally in US equities has also been consistent with this theme of recession denied. The US equity risk premium is at lows not seen since 2007, implying that there is currently very little compensation for risk taking:
John Authers noted in his half-year review:
The key differentiator between bulls and bears is not the economy, but rather about liquidity and monetary conditions. And this leads to a final reason why the first half of 2023 has been so good for those who invest in the S&P… Monetarists, and those who believe that inflation has been driven more by factors like supply bottlenecks than by the money supply, suspect that higher interest rates are going to inflict an unnecessary recession.
As noted in our book club series on his latest book, Sumner is a pragmatist who maintains we should judge the stance of monetary policy based on macroeconomics outcomes, not what we assume those outcomes should be based on misconceptions about the stance of monetary policy. If the US economy and equity market have surprised on the upside, it quite likely that the effective stance of monetary policy is not as tight as many have assumed. That in turn implies that monetary policy has more work to do. But as we have seen in the case of labour market, we shouldn’t neglect the potential contribution of the supply-side (and supply-side policy) in moderating inflationary pressures. The NAIRU is partly a policy choice.
Meta’s Threads as a victory for federalism
Threads, Meta’s new microblogging platform, runs on the ActivityPub protocol that also underpins Mastodon. This means that Mastodon and Threads are potentially interoperable, with users able to follow each other and exchange messages across platforms. Early days, but holds out the promise of transforming the social web into an open, multi-jurisdictional, distributed order. Federalists will appreciate the possibilities.
I won’t be signing-up for Threads just yet. For now, you can find my short-form content, such as it is, on Substack’s Notes. This is a short primer I wrote in how to use Notes.
ICYMI
Who killed Google Reader? Ten years after its untimely death, the team that built the much-beloved feed reader reflects on what went wrong and what could have been.
Federal cabinet to decide on the next RBA Governor in the next two weeks, with Treasury Secretary Steven Kennedy the front-runner, according to The Australian.
Zac Gross takes a deep-dive into the RBA staff survey, highlighting the shocking disconnect between RBA research and monetary policy decision-making.