The United States: Credit squeeze increases the risk of recession

Fed key rate hikes are starting to bite in financial markets and the economy. Because of tighter lending conditions for SMEs and more cautious households, the US will enter a mild recession in 2023. Late-cyclical inflationary forces in the labour market will offset falling goods prices. Core PCE inflation will not revert to target until 2024. The Fed will start rate cuts in late 2023, once the labour market has clearly weakened.

The window for a soft landing has shrunk. GDP growth dampened in Q1, due to a large inventory correction. Underlying demand was strong, however, challenging the Federal Reserve’s efforts to cool the economy to a level that is sustainable in the long term. Sticky inflationary forces − with upwardly revised CPI data at the end of 2022 − and a hot labour market are creating pressure for further key rate hikes. Meanwhile tighter credit conditions are likely to cause the market to do part of the job for the Fed. The need to agree on raising the federal debt ceiling adds to the uncertain outlook.

Key data

Year-on-year percentage change

*% of labour force **% of GDP, ***Upper end of the Fed’s range.    Source: Macrobond, SEB

Our main scenario remains a mild recession, with GDP ending up a bit below its year-end 2022 level by the close of 2023. We believe that strong consumer demand early this year reflects temporary factors, including abnormally warm weather during January-February and a large one-off social security payment in January. We still expect several quarters of declining GDP, but our full-year 2023 forecast is above zero – and raised to 0.7 per cent from 0.5 per cent in January’s Nordic Outlook. We are lowering our 2024 forecast from 1.2 to 0.9 per cent.

Pessimistic companies and households. Sentiment surveys support a loss of momentum. The purchasing managers’ index for big manufacturing firms (ISM) is at levels that have historically often been followed by recessions. The service sector showed strength early in 2023 but declined in March. Consumer sentiment bottomed out when inflation peaked last summer but has remained weak, according to the Michigan survey.

Credit crunch despite signs of stabilisation

For a few weeks in March, it looked as if the collapse of California’s tech-oriented Silicon Valley Bank (SVB) might trigger a new US banking crisis. Government actions – guarantees for all deposits in SVB and another crisis-hit bank, reassuring statements from the White House and new Fed borrowing facilities − appear to have stemmed an acute crisis of confidence, at least temporarily. Bank deposits and demand by banks for emergency loans from the Fed have stabilised. The regional banks that have been the focus of concern are included among statistics for large banks, which may conceal flows within this category. But so far, overall data show neither a general exodus beyond the previous downward trend nor an acute credit crunch.

Higher borrowing costs. Yet we expect that events will reinforce the already ongoing credit squeeze. The consequences of excessive risk-taking tend to surface during rate hiking cycles, and we are likely to see more trouble spots. The fundamental problems that triggered the collapse of SVB are found − though in less extreme form − at other banks: large uninsured deposits that face competition from higher yields on Treasury bills and money market funds, as well as large unrealised losses on government and mortgage bond holdings as yields have climbed. The Fed’s new one-year loans, secured by government and mortgage bonds, reduce the risk of acute liquidity crises but are expensive. Bank deposit rates are thus likely to be raised to stop further outflows. The bill for insuring deposits at SVB is being paid by other banks, via increased contributions to the Deposit Insurance Fund. Smaller banks run the risk of being subjected to stricter regulation and supervision.

Small and medium-sized enterprises (SMEs) are the most vulnerable. Indices of financial conditions have not moved very much, and credit spreads have narrowed again. However, credit conditions for SMEs, which depend on banks for their funding, are being tightened. According to a survey by the National Federation of Independent Business (NFIB), in March credit availability and expected credit conditions were the tightest since late 2012 and were at levels normally seen during economic downturns.  We believe this will lead to a downturn in business investments, which is supported by weak new order assessments (ISM) and a slowdown in factory orders. A bottoming-out trend for housing activity is also being challenged by tighter lending conditions. We expect residential investments to continue falling during the first half of 2023.

Pandemic savings buffers are emptying 

Large post-pandemic household savings buffers caused consumption to defy both falling real incomes and higher interest rates. More than half of these buffers had probably been emptied by end-2022. We believe they will be completely exhausted by end-2023 and increasingly concentrated among affluent households with lower consumption propensities. But pay and employment growth will provide new support. Real wages are no longer falling. Households are also still spending a larger-than-usual share of income.

We believe that households will become more cautious as the labour market starts to slow down more clearly. We expect such a trend to begin this spring and lead to falling consumption during the second half. But due to the strong start to the year, we are raising our full-year forecast for private consumption from 0.9 to 1.2 per cent, while lowering it to 0.8 per cent in 2024.

Resilient service jobs

The number of new jobs is still growing faster than the pre-pandemic average of around 180,000 jobs per month and the 100,000 or so the Fed views as compatible with labour market stability. Over the past three months, job growth has been dominated by health services, leisure/ hospitality and the public sector, with the last two areas not yet having regained pandemic losses. Looking ahead, we believe that hiring in these sectors will also decelerate, contributing to a general slowdown. But the current economic cycle, being led by non-interest rate-sensitive service sectors, has probably dulled and delayed the impact of rapid Fed rate hikes.

Early signs of normalisation. Unemployment is expected to start rising this summer from today’s 50-year low of around 3.5 per cent, peaking at over 4.5 per cent in 2024. Rising lay-offs, and a slow increase in initial jobless claims, support expectations of a gradual labour market cooldown. The number of job openings is falling faster, from high levels, while small businesses have revised hiring plans below pre-pandemic levels. Meanwhile the supply side has improved. Labour force participation for prime ages 25-54 is back at early-2020 levels. Immigration has recovered from Trump’s restrictions and pandemic border closures. Overall participation remains lower, but this reflects an ageing population. Labour supply would probably have undergone a declining trend even with no pandemic.

Pay hikes continue to signal a tight labour market. Hourly earnings growth dampened during the first quarter and is approaching levels broadly consistent with the 2 per cent inflation target. According to the more reliable employment cost index, ECI, wages continued to rise at an annualised pace of nearly 5 per cent, however.  

Hard-to-assess Philips curve a Fed dilemma

The period 2000-2019 was characterised by a flat Philips curve (the relationship between unemployment and inflation), in which inflation was stable at low levels over an economic cycle. Inflation became entrenched at high levels after the pandemic, which may indicate that the non-accelerating inflation rate of unemployment (NAIRU) has shifted upward. If so, 5-6 per cent unemployment may be needed to return to the 2 per cent inflation target. But another interpretation is that the relationship is stronger, and the Philips curve steeper, when the labour market is very tight. If so, the Fed may not need to cool off the economy as much. We are leaning towards the latter hypothesis, which is also supported by data. The Philips curve’s uncertain shape is likely to make the Fed less inclined to react pre-emptively to signs of a weaker economy until it sees clear evidence that inflation is on its way back to target.

Sluggish service prices a Fed headache. Total inflation has fallen fast as energy prices have decreased. Other goods prices have also declined due to normalised supply chains, falling freight rates and shrinking wholesale and retail margins. Falling used car prices − following a sharp rise during the pandemic − have also been important for the decline. But the Fed has shifted its focus to service prices excluding energy and rents. According to the Fed, this metric best captures impulses from wages and underlying inflationary pressures in the economy. It has stayed largely flat at uncomfortably high levels around 4.5 per cent since mid-2022. The lack of progress in lowering underlying inflation is one reason why the Fed continued rate hikes in March despite the SVB crisis, and why almost all Fed policymakers stuck to their plan for a further hike. Despite easing pay increases, our forecast suggests that service inflation will remain a headache for the Fed this year.

No key rate cuts until late 2023. Still, the Fed can hardly ignore a situation of falling employment and rising joblessness, which is what we expect to see in the second half. We are therefore sticking to our forecast of a final hike in the federal funds rate to 5.00-5.25 per cent at the meeting in early May. As before, we believe that the Fed will start cutting during Q4. We expect the key rate to stand at 4.50-4.75 per cent by the end of 2023 and at 2.75-3.00 per cent by end 2024. We think it will be hard for the Fed to separate the slimming of its balance sheet (quantitative tightening, QT) from its rate decisions and expect that QT will end next year.

Debt ceiling negotiations deadlocked. The inability of congressional Democrats and Republicans to agree on an increase in the federal debt ceiling is a new threat to financial markets and the US economy. The estimates of the Congressional Budget Office (CBO) indicate that the federal government will run out of cash between July and September 2023. A decline in tax revenues means that it may happen as early as June. Congress has three options: raise the ceiling, suspend it temporarily or postpone the problem by raising it slightly.

Historically, a last-minute agreement has always been reached, avoiding a self-inflicted US default. However, increased polarisation between and within the parties has increased the risk that something will go wrong. Republican House Speaker Kevin McCarthy has proposed a 10-year savings plan in exchange for a one-year raise in the debt ceiling, but the Democrats do not want to face a new debt ceiling crisis during the 2024 presidential election campaign and are opposed in principle to the debt ceiling being used as a bargaining weapon. Demands for drastic cuts next year and slashing large parts of President Biden’s climate agenda, the Inflation Reduction Act, IRA, are hard to stomach.

Starvation diet or panic cuts? A long-term austerity policy like the one that President Obama was forced to accede to in 2011 could limit the federal government’s ability to help sustain the US economy in case of a new downturn. But a failure to raise the debt ceiling would be much worse. The latter option would create demands for immediate and drastic spending cuts, including defence, health care and pensions, given a deficit of over 5 per cent of GDP. However, we believe that the US Treasury has the tools to prioritise interest payments on the federal debt. In sum, the budget process needs to be reformed to address both persistent deficits and underfunded social security systems. But the latter are politically sensitive and do not lend themselves to a “game of chicken” over America’s willingness and ability to make good on its obligations.