Small changes can indeed have a significant impact. This was exemplified by the Consumer Price Index (CPI) report for March: core inflation, which excludes volatile components such as food and energy, rose by 0.36% month-on-month (MoM), a slight increase from the previous 0.34% MoM. However, the U.S. Bureau of Labor Statistics adheres to the convention of rounding to the nearest tenth. As a result, March was reported as a 0.4% MoM increase, unsettling markets as it marked an unexpected rise – expectations had been set at a 0.3% MoM increase. This led to further shifts in market expectations regarding the future path of policy rates.

At the start of the year, markets were anticipating approximately six 25-basis-point cuts within a 12-month timeframe. These expectations roughly halved in the weeks leading up to the release of the CPI number for March, and following the release, markets are pricing in fewer than two rate cuts until the end of the year. Simultaneously, the surprise pushed out the expected timing of the first rate cut from June to later this year. Moreover, this shift in expectations sparked a lively debate about whether the Federal Reserve (Fed) will be able to cut policy rates at all, or even if it needs to increase them once more to curb inflation. The Personal Consumption Expenditures (PCE) price inflation, which is arguably more relevant for monetary policy and is released two weeks after the CPI report, was back in line with expectations but that did not alter market sentiment.

Even a robust but softer-than-expected first quarter GDP print was interpreted on the hawkish side, since domestic demand remains strong and is only slowing somewhat from last year. The robust growth and somewhat stagnant disinflation were both acknowledged by central bankers at the most recent FOMC meeting. However, they appeared to maintain a slightly dovish bias despite those seemingly hawkish inputs. Moreover, following Fed Chair Powell’s remarks, it became clear that the threshold for further rate increases remains high as “the totality of data” needs to go in the wrong direction – not just disinflation stalling by one particular measure.^[1] Powell reiterated that there is no urgency to adjust policy rates, as they are comfortable with the current monetary policy setting. They believe it is sufficiently restrictive and they are willing to wait for more data to determine whether their confidence in the disinflationary process is reinforced.

Upon examining inflation, we indeed observe a shift in momentum as measured by the 6-month inflation rates. The latter half of 2023 was particularly characterized by a significant deceleration in 6-month inflation rates, even dipping slightly below 2 percent on an annualized basis in November and December. However, the first quarter has so far indicated a re-acceleration. Chart 1 illustrates this for both the core CPI and the core PCE measure.

Chart 1: The process of disinflation stalled recently.

^{Sources: Haver Analytics, BEA, DWS Investment GmbH, as of May 2024}

This increase in momentum did not necessarily elevate the year-over-year inflation figure, but it also did not decrease it further. From this perspective, the process of disinflation has come to a halt. However, a significant divergence becomes apparent when comparing both measures. While the core CPI indeed accelerated further in 6-month terms, sending hawkish signals, the PCE itself (unlike the six-months annualized measure) stopped short of such an acceleration, somewhat alleviating the pressure to react hawkishly. But, which one to trust?

The primary distinction between the CPI and the PCE lies in how the basket of goods and services considered is constructed. For the CPI measure, the weights of goods and services are adjusted annually based on consumption patterns from the previous year, resulting in a 24-month lag before changes in spending are reflected in the index.^[2]By contrast, the weights for the PCE measure are updated quarterly, thereby quickly reflecting substitution effects when consumers shift spending from higher-priced goods to lower-priced ones.^[3]The advantage of this dynamic approach is its higher accuracy in measuring inflation and spending trends, making it the preferred measure for economists and central bankers. However, there's more to it. While the CPI is survey-based and only considers what urban consumers report they have consumed, the PCE takes into account the actual spending of all consumers, both urban and rural, thus providing a more comprehensive picture of the economy.^[4]These differences are reflected in the sometimes significant disparities in weights for certain categories. For instance, the cost of shelter accounts for roughly 30 percent of the CPI and "only" about 15 percent of the PCE. Certainly, in urban areas, the cost of housing constitutes the largest share of monthly consumption, but this is not the case when considering the economy as a whole.

Post-pandemic, shelter prices have been a constant source of concern for economists. Typically, the shelter component in CPI (or housing in PCE) follows the various available house-price and rent indices with a robust time lag of 12 months (Chart 2).

Chart 2: Costs of housing set to decline further – at least for some time.

^{Sources: Haver Analytics, DWS Investment GmbH, as of May 2024}

At present, this relationship appears to hold, and we anticipate that it will continue. However, this relationship also suggests that the process of disinflation from this component will likely require more time, maybe increasing the wedge between CPI and PCE even further. Also note that as chart 2 shows, house prices as measured, especially by the S&P CoreLogic Case-Shiller index have lately rebounded, though the same is not true for rents, which we would argue is the more relevant measure for the CPI Shelter component. Another closely monitored part of inflation is the so-called core-services excluding housing/shelter, often referred to as super-core inflation, which accounts for another significant portion of the overall PCE (approximately 46 percent). Right now, the fate of these prices is likely closely linked to labor market tightness (see Chart 3).

Chart 3: Labor market tightness^[5] vs. core service prices ex. shelter

^{Sources: U.S. Bureau of Economic Analysis, Haver Analytics, DWS Investment GmbH as of May 2024}

When demand for labor exceeds supply, labor market tightness begins to increase, which corresponds to negative values on the horizontal axis of the chart. Prior to the pandemic, inflationary pressures from labor markets for super core PCE inflation were relatively subdued. During the pandemic, we first observed a lack of demand core-services excluding housing/shelter due to lockdowns, which corresponded to lower inflation rates. This was followed by a surge in labor demand as economies reopened, especially for labor intensive services such as hospitality and leisure. At the same time, we observed a shortage of labor supply due to limited migration and generous fiscal support that likely kept some people out of the labor force. As a result, inflationary pressures increased significantly. Once the vacancy rate (our measure for labor demand) peaked around April 2023, inflationary pressures stabilized at high levels and began to slowly decline around the time the Fed started to raise rates. The chart suggests that this process of disinflation is now in an advanced stage but still not complete; pre-pandemic levels of inflation in core-services excluding housing/shelter have not yet been reached, making it harder to reach the overall target of 2% in PCE inflation, the Fed’s preferred measure of success. However, increased migration provides hope for a further increase in labor supply and therefore the possibility of lower inflation rates over the medium term. The Congressional Budget Office (CBO) recently updated its demographic outlook, significantly revising its immigration estimates upward (see Chart 4).^[6]

Chart 4: CBO now estimates a significant uptick in immigration

^{Sources: Congressional Budget Office, DWS Investment GmbH, as of May 2024}

For 2023 alone, the influx of people is now estimated at 3.3 million, compared to roughly 1 million previously. This could, at least in parts, explain the upside surprise in personal spending we have seen in the past as well as the stalling of disinflation right now. Given that consumption needs to be earned first, at least in most cases - and not every dollar earned is spent right away or even over the medium run - the expectation of slowing consumption should still hold, however. Therefore, the potentially permanent increase in labor supply due to higher immigration could help alleviate some of the tightness in the labor market and should be net positive for the disinflationary process looking ahead.

As of now, however, it seems like labor markets are performing exceptionally well. The unemployment rate rests at 3.8 percent and hiring – in form of gains in non-farm payrolls – with an average of 276k in the past three months reminds experienced observers more of a small boom than signalizing a cooling of the economy. But migration also plays its part here. A recent study suggests that with such high levels of migration, levels of employment gains in the area of 160k to 200k would be sustainable in the sense of not adding to inflationary pressures, compared to 60k to 100k neglecting the effects from migration.^[7]

Moreover, the current low unemployment rate might overstate the health of the economy somewhat. Applying the Sahm Recession Indicator to unemployment rates of various states gives the impression that around 20 of them are close to entering one.^[8] While the inventor of the indicator herself warned of applying the rule to individual states, we find the insights useful to track the underlying dynamics.^[9] And these underlying dynamics might just get more important for the Fed as well. At the May FOMC meeting press conference, Fed Chair Powell shifted the focus somewhat back on the state of the labor markets, away from inflation. The reason for this was not only because inflation is now closer to the Fed’s target but also that a deterioration of labor market conditions might be one trigger to cut rates.^[10] Our interpretation to get somewhat cautious on labor market conditions rests on the fact that once the unemployment rate starts to increase, it historically does so with increasing speed that seems hard to tame. From a monetary policy perspective, it is therefore preferable to prevent from such a dynamic early on. Keeping rates higher for longer – as now intended by central bankers – per definition increases the probability of those so-called unintended consequences from past rate hikes.

And there is a lot of evidence that higher policy rates are already taking a toll on the economy. Personal interest payments as share of disposable income, for instance, remains high and close to levels last seen just before the global financial crisis at around 2.6 percent. On top of that, we observe that delinquency rates for consumer loans past due 30 days are already somewhat higher than before the pandemic, which fits with a decline of the savings rate. Both typically indicate that consumer’s ability to spend will slow down looking ahead. It seems like, with the depletion of excess savings out of the pandemic, people indeed rely more on labor income. With an easing of labor market conditions, we believe that wage growth will normalize, too. And while migration supports higher supply of labor, there are ongoing indications that demand for labor will cool further as well. The biggest employer in the U.S., small businesses, already indicate a harder time to meet its financing needs that go along with muted sales expectations and declining job openings.^[11]

We continue to believe there is a residual risk that past rate hikes might slow down the economy more than currently anticipated, but we concede that the likelihood of this happening is much lower now compared to a few months ago. Despite this optimism, a formerly very reliable recession indicator – an inversion of the Treasury yield curve – has not yet normalized, and the question remains whether this tool still deserves its place in our toolbox. Typically, an inverted yield curve is defined as a situation when yields of longer-maturity Treasuries fall below the yields of shorter-term maturities, e.g., the spread between the 10-year and the 2-year Treasury yield turning negative (see Chart 5).

Chart 5: Treasury Yield Curve still inverted

^{Sources: Haver Analytics, DWS Investment GmbH, as of May 2024}

It’s important to note that while an inverted yield curve has been a reliable precursor of recessions in the past, it doesn’t necessarily lead to a recession. Nor can we draw any conclusions regarding the length or depth of a following recession in relation to the length or depth of the inversion. It's merely a signal of market sentiment and economic conditions.^[12]However, some research suggests that an inverted yield curve might do more than predict a recession as it might contribute to tightening economic conditions by leading banks to reduce the supply of loans.^[13]

The question remains: Is this time different? Once the yield curve normalizes, will we indeed slip into a recession or can a soft landing be achieved? What sets this period apart from past cycles is the policy response to the external shock known as the pandemic. The massive fiscal and monetary response might be the closest realization of Ben Bernanke's interpretation of Milton Friedman's concept of "helicopter money".^[14] Back in 2002, such a "money-financed tax cut" was discussed in the face of deflationary threats. At this juncture, the policy measures are believed to have inadvertently engineered an economic cycle of their own, culminating in a supply-demand imbalance and consequently, elevated inflation rates. Not too long ago, the Federal Reserve appeared ready to respond to the cooling of this cycle. However, an unexpected surge in immigration, or perhaps a rebound in labor productivity, may prolong it. This could potentially delay the necessary adjustments to the monetary policy rate, which are crucial for ensuring a soft landing, until later in 2024.

Overview: key economic indicators

	2023				2024				2025
	Q1	Q2	Q3	Q4	Q1	Q2F**	Q3F**	Q4F**	Q1F**
GDP (% qoq, annualized)	2.2	2.1	4.9	3.4	1.6	0.0	0.6	0.8	2.0
Core inflation (% yoy)*	4.8	4.6	3.8	3.2	2.9	2.6	2.4	2.4	2.3
Headline inflation (% yoy)*	4.4	3.3	2.8	2.6	2.6	2.6	2.5	2.4	2.3
Unemployment rate (%)	3.5	3.6	3.7	3.7	3.8	4.0	4.0	4.0	4.0
Fiscal balance (% of GDP)				-6.5				-6.2
Federal funds rate (%)	4.75 - 5	5 - 5.25	5.25 - 5.5	5.25 - 5.5	5.25 - 5.5	5 - 5.25	4.75 - 5	4.75 - 5	4.5 - 4.75

*PCE Price Index

** Forecast

^{Source: DWS Investment GmbH as of 3/5/24.}

Forecasts are not a reliable indicator of future returns. Forecasts are based on assumptions, estimates, views and hypothetical models or analyses, which might prove inaccurate or incorrect.