The US economy is sending mixed signals

We made some key judgements in our latest global forecast update.

  • The first was to push back the timing of recession to the turn of the year based on resilient activity.
  • The second reaffirmed that we don’t see the labor market cooling in a way that would open the door to a soft landing.
  • And relatedly, that sticky inflation pressures would force the Fed to raise rates further in coming months.

However, data are not providing consistent signals, and some indicators suggest the economy could be headed in different directions. In this note we examine the data that challenge our key judgements and set out waymarks which might tell us it is time to change tack.

Base case 1

The economy is chugging along

Short-term US growth has been more resilient than expected, delaying a recession until the turn of the year. Solid consumer spending, supported by growing real incomes and strong balance sheets, underpins this trend. A cross check with the labor market shows hiring is still robust and, while business investment has been weak, there are signs of stabilization in the sectors most affected by the Fed tightening policy, like residential investment.

Or is it?

It is not uncommon for downturns to catch forecasters by surprise, and recession-dating committees can benchmark its start ahead of the first contraction in quarter-over-quarter Gross Domestic Product (GDP). Some data are showing signs of cracking. If we look at Gross Domestic Income (GDI) we are already in a recession. This measures aggregate activity via income received – e.g., profits and wages – as opposed to GDP, which looks at spending or production. GDI has been negative in three of the last four quarters and is already down 1.4% from its peak, consistent with a decent-sized recession. Corporate profits, which softened even further in Q1, contracting over 5% annualized on the quarter, have caused this weakness.

Figure 1: The gap between GDP and GDI is widening

Source: BEA, Haver, June 2023

GDP generally garners more attention given its relative timeliness, however, GDI historically has smaller revisions and research has shown that, along a number of dimensions, GDI may be a better measure of economic strength and business cycle turning points.

Figure 2: The Conference Board Leading Indicator is consistent with a recession

Source: Conference Board, BEA, Haver, abrdn, June 2023

In terms of higher frequency data, the Conference Board Composite Leading Indicator – which is comprised of a range of labor market, sentiment survey, investment, housing and financial indicators – has declined for 13 straight months and is sitting at levels consistent with a recession (see Figure 2). In part this will be picking up the widespread tightening in bank lending standards over recent quarters, which again would be consistent with a downturn historically.

What to watch:

  • Signs of corporate retrenchment on the back of weak profits and tight credit conditions (core durable goods orders, hiring trends, credit growth).
  • Consumer spending indicators – the latest retail sales data were soft, with downward revisions to previous months. It will be important to see how this reflects on broader services spending.

Base case 2

Resilient activity is not compatible with a soft landing

We don't think a recession is imminent, but we do think one is coming, and indeed required to resolve the imbalances in the economy. The labor market remains far too tight, driving wage growth well above the level consistent with the Fed's inflation target. Importantly, we don't believe a large recovery in labor supply will resolve these tensions, nor that an improvement in labor market efficiency will allow labor demand to soften without surging layoffs.

Or is it?

The labor force shrank during the pandemic, but has more recently rebounded back to its pre-COVID trend, helped by stronger immigration and the return of workers who were sitting on the side lines, likely due to attractive wage growth.

We question how much further this recovery has to run as an aging population weighs on the labor force. But further strong increases in labor supply would help loosen the labor market, as would some moderation in labor demand, and there have been signs this is softening. Hours worked are growing more slowly (see Figure 3) and vacancy rates have fallen from their peaks. Similarly, initial unemployment insurance claims have moved higher, quits are slowing and the Conference Board labor market differential has moved back to more normal levels, signaling workers are less confident they will find a new job quickly. If these trends were to continue, absent a large rise in layoffs, this would be a sign that the labor market is cooling in a benign manner.

Figure 3: Signs of easing labor demand?

Source: BLS, Haver, abrdn, June 2023

We can monitor progress towards a benign cooling in labor demand through the Beveridge Curve, which tracks vacancy rates versus unemployment rates. The outward shift in this curve during the pandemic (see Figure 4) represented a decrease in matching efficiency, as firms struggled to find workers to fill their openings. There has been a modest fall in vacancies relative to unemployment since, which implies some recovery in the matching dynamic required to allow labor demand to cool without a rise in unemployment. But clearly this has a long way to run before we see a return to more normal pre-pandemic labor market dynamics.

Figure 4: Heading in the right direction

Source: BLS, Haver, abrdn, June 2023

Finally, if we focus on average hourly earnings (AHE), there is evidence that the combination of better labor supply and demand is helping ease wage pressures with this now down to 4.3% year over year. But it is important to note that compositional issues might be driving this, with strong jobs growth in lower paying services, perhaps artificially depressing labor costs (see Figure 5).

What to watch:

  • Participation rates and population growth for signs of a further acceleration in labor supply;
  • Job openings relative to unemployment, for signs that cooling labor demand is not triggering layoffs;
  • Alternative measures of wage growth including the Atlanta Federal Reserve's wage growth tracker and the Employment Cost Index (ECI);
  • Layoffs, hours worked, and initial claims will give us a good read on how businesses are changing hiring patterns.

Figure 5: Consider a range of wage measures

Source: Atlanta Fed, BLS, Haver, abrdn, June 2023

Base case 3

Core inflation remains sticky

The annual rate of headline inflation has more than halved from its peak last summer, but progress in core inflation has been slower, driven by stickier services prices. Our favorite measure of these underlying inflation pressures is core services inflation excluding rents and health insurance, with the latter severely depressed at present due to a change in methodology at the BLS. This measure shows limited progress back towards the Fed’s target (see Figure 6).

Or is there?

We do have to be careful when focusing on bespoke exclusion measures of inflation, as these can be constructed in ways to match our prior expectations. More formal statistically derived exclusion measures, like the weighted median – which looks at the rate of inflation of the item at the middle of price changes in the basket – or the trimmed mean – which strips out both the largest price increases and decreases – signal a somewhat slower pace of inflation in recent months (see Figure 8).

Figure 6: Inflation persistence depends on the measure

Source: BLS, Haver, abrdn, June 2023

More fundamentally, the weak trend in AHE is also a positive one (if accurate) for the future path of services inflation. Indeed, the decline in measured compensation growth, and less bad productivity growth, have helped slow unit labor cost growth (ULC) for firms somewhat. This could signal that pipeline inflation pressures stemming from the labor market may begin to wane more markedly (see Figure 7). However, these data are very volatile.

Figure 7: Revisions to ULC are a more positive signal on the inflation front

Source: BLS, BEA, Haver, abrdn, June 2023

What to watch:

  • It is best to focus on a wide range of measures of inflation, but we think our preferred indicator, which strips out housing costs and health insurance, is the closest to the Fed’s preferred measure of core PCE excluding shelter.
  • Inflation expectations, given the influence of these on the shape of the Phillips curve and to understand the relationship between economic slack and inflation. Our recent analysis suggests expectations can often be backward looking following periods of elevated inflation. We will be watching these closely to understand the extent to which this may add to inflation stickiness.

Figure 8: Tracking an array of inflation measures is important

Source: BLS, Haver, abrdn, June 2023

A scenarios-based approach is best amid data uncertainty

The alternative interpretations of the current data flow outlined in this note highlight some of the very plausible scenarios that could play out instead of our base case. We could be too optimistic, meaning a downturn is already upon us. This could even cast doubt on the final Fed hike we have penciled in for July.

If we are right about the economy being resilient, there are still a range of ways the macro story could play out differently. A later recession could provide more time for labor market imbalances to resolve in a benign manner, widening the path for a soft landing.

Alternatively, that same economic resilience, absent further loosening in the labor market, could raise the risk of our “two bites of the cherry” scenario materializing if it persists into 2024. This would force the Fed to significantly tighten policy further, leading to a deeper and more protracted recession. All of this highlights the importance of monitoring incoming data very closely, and in particular, the waymarks we have identified, for signs that other scenarios are becoming more or less likely.

Key takeaways

  • The wealth of information available to track the state of the US economy is both a blessing and a curse – particularly when these data provide mixed signals.
  • Our best reading of the current data flow is that the economy is not in recession (or on the verge of one), that the labor market remains unhealthily tight and inflation stubbornly strong.
  • However, some data contradict these judgements. Gross Domestic Income suggests we are well into a downturn, and some higher frequency activity measures, and bank lending standards, are consistent with recession.
  • Similarly, unerring strong payrolls data aside, softening quit rates, weak household survey employment, slowing hours worked, a modest decline in vacancies and stronger labor supply could all tell us that the labor market is loosening in a manner more consistent with a soft landing.
  • Finally, inflation has been falling on the back of lower food and energy price pressures, and some exclusion- based measures suggest more progress in cooling core style measures of underlying inflation.
  • This uncertainty means it is critical to monitor incoming data extremely closely and lean heavily on scenarios when assessing the outlook.
  • Our base case is for the US economy to enter recession around the turn of the year. But, if activity remains resilient, there is a still a path to a soft landing based on inflation cooling in a benign manner. If inflation does not cooperate, expect the Fed Funds rate to peak at a level above 6%.