Another strong personal income and spending report, but beware the uptick in inflation

 

 - by New Deal democrat



Personal income and spending has become one of the two most important monthly reports I follow. This is in large part because the big question this year is whether the contractionary effects of Fed tightening have just been delayed until this year, or whether the fact that there have been no rate hikes since last summer mean that the expansion will strengthen.

Because real personal spending on services for the past 50 years has generally risen even during recessions, the more leading components of this report have to do with spending on goods. Additionally, there are several components that form part of the NBER’s “official” toolkit for determining when and whether a recession has begun, including real spending minus government transfers, and real total business sales. 

Let’s look at each of them in turn. Note that in all graphs below except for YoY comparisons, and the personal saving rate, is normed to 100 as of just before the pandemic.

Real income and spending

In March, nominally income rose 0.5%, while nominal spending for the second month in a row rose a sharp 0.8%. Prices as measured by the PCE deflator increased 0.3% for the month, meaning that in real terms income increased 0.2% and spending rose 0.5%. Since just before the pandemic real incomes are up 7.2%, and spending is up 11.3%:

On a YoY basis, the PCE price index is up 2.7%, the second monthly increase in a row from January’s three year low of 2.4%. While n the previous 16 months, the YoY measure had been declining at the rate of 0.25%/month, suggesting that it would hit the Fed’s 2.0% target this spring, that trend has stopped:

The issue going forward is going to be if this was a temporary pause, a stalling of progress towards the Fed’s goal of 2.0%, or an outright reversal of trend. Suffice it to say, the Fed is not going to cut rates in the near future under either of the two latter scenarios.

As I indicated in the intro, for the past 50+ years, real spending on services has generally increased even during recessions. It is real spending on goods which declines. Last month real services spending rose 0.2%, while real goods spending rose a sharp 1.1%:

Further, real durable goods spending tends to turn before non-durable goods spending. The former rose 0.9% for the month, and the latter 1.3%. Both of these have now totally or almost totally reversed previous months’ sharp declines:

Savings

Another important metric for the near future of the economy is the personal savings rate. In February it declined a sharp -0.5%. In March it declined another -0.4% to 3.2%. This longer term look shows how the present compares with the all time low rate of 1.4% in 2005:

On the positive side, the declining trend in this rate since last May indicates a lot of consumer confidence. But on the negative side it is close to its post-pandemic low of June 2022 of 2.7%, as well as its previous all-time low range, indicating vulnerability to an adverse shock. One of my forecasting models uses such a shock as a recession warning indicator. In any event, there is no such shock indicated at the moment.

Important coincident measures for the NBER

Also as indicated above, the NBER pays particular attention to several other aspects of this release. Real income excluding government transfers (like the 2020 and 2021 stimulus payments) rose 0.2% for the month, continuing its consitent increasing trend since November 2022:

Finally, the deflator in this morning’s report is used to calculate, with a one month delay,  real manufacturing and trade sales. This rose 0.5% in February, but is still below its November and December 2023 readings:

Summary

I had described January’s report as being pretty close to “Goldilocks,” and last month’s as something of “anti-Goldilocks.” In March we were closer to Goldilocks than not. Real Income, including real income less government transfers, and spending both rose, and spending on goods rose sharply. Real manufacturing and trade sales also rose. The saving rate shows near term confidence, but is a longer term concern. Aside from that the only significant negative was the second straight increase in YoY PCE inflation.

The consumer remains healthy, but is digging into their savings to support much of the increase in spending. In addition to the price of gas, which has recently been rising again, but if this renewed sharp increase in spending is feeding through into a renewed increase in the inflation rate as well (and it may well not be), that would not be good. Because if that’s the case, the Fed is going to consider not just not lowering interest rates, but possibly another increase as well. 

Leading indicators in the Q1 GDP report are mixed

 

 - by New Deal democrat

Most of the commentary you will read about Q1 GDP that was released this morning will be about the core coincident components. For that I will simply outsource to Harvard’s Prof. Jason Furman:

“much of the slowdown was in non-inertial items like inventories (-0.35pp) and net exports (-0.86pp). The better signal of final sales to private domestic purchasers was 3.1%.”

I agree.

With that out of the way, as usual, my focus is instead on what the leading indicators contained in the report can tell us about the months ahead. There are two such long leading indicators: private residential fixed investment (basically, housing) as a share of GDP, and deflated corporate profits.

Let’s look at each one in turn.

Nominal private residential fixed investment increased 3.5% during the first quarter. Real private residential fixed investment increased 3.3%. Since both of those were bigger increases than the % change in nominal and real GDP for the quarter, respectively, both improved as a share of GDP, as shown below:

Not strong improvement, but improvement nevertheless. 

Since corporate profits for the Quarter aren’t reported in the first release, we turn to the proxy of proprietors’ income. The proper measure deflates by unit labor costs, but those won’t be reported until later either, so the GDP price deflator is a good proxy. Going back 50+ years, our substitute typically turns coincident with or one quarter later than the official leading indicator.

Nominally proprietors’ income rose 0.5% in the quarter. But since the GDP deflator increased 0.8%, their deflated income declined slightly (blue in the graph below):

Real proprietors’ income has been essentially flat for the past two years.

It’s worth noting that (nominal) corporate profits as reported to Wall Street, updated through last week, with the exception of Q3 of last year haven’t gone anywhere in two years either:

When profit growth stalls, companies start looking around for costs to cut, and usually that includes employees. So that’s not good.

So our leading indicators from the Q1 GDP report score as one positive and one neutral. Several other of the long leading indicators have gotten “less bad” in the past half year. Once Q1 bank lending conditions are reported in a week or two, I will update my top-of-the-line long leading forecast through the end of the year.

Jobless claims continue their snooze-fest

 

 - by New Deal democrat

[Note: I’ll put up a post discussing Q1 GDP later today.]

Initial and continuing claims continued their snooze-fest this week.

Initial claims declined -5,000 to 207,000, continuing their nearly 3 month long range of between 200-220,000 per week. The four week average declined 1,250 to 213,250. This average has remained in the 200-225,000 range for over half a year! Finally, with the typical one week delay, continuing claims declined -15,000 to 1.781 million:

As per usual, for forecasting purposes the YoY range is more important. Here, initial claims were down -1.0%, the four week average down -1.8%, and continuing claims higher by 3.4%, still the lowest comparison for continuing claims since February 2023:

Needless to say, this is potent evidence that we can expect the economy to continue to expand in the next few months.

As you also might expect, with initial claims lower YoY for the month of April so far by -2.9%, this also suggests that the unemployment rate will trend lower YoY over the coming months, towards 3.7% or even 3.6%:

Since initial claims (and to a lesser extent continuing claims) lead the unemployment rate, the Sahm rule for recessions is not going to be triggered.

In addition to housing, manufacturing is range-bound as well

 

 - by New Deal democrat

First off, let me reiterate that my focus this year is on manufacturing and construction. That’s because these are the two sectors the waxing and waning of which have almost always determined if the US economy is growing or not. By contrast, for the past half century or more the production and consumption of services has tended to increase even right through most recessions.

With that framework in mind, yesterday I wrote about how, following interest rates, housing is range-bound.

This morning durable goods orders for March were reported, which gives me a good opportunity to update the state of the maufacturing sector.

Total durable goods orders rose 2.6% month over month. Core capital goods orders rose 6.0%. These series are very volatile. Thus the big increase in orders was only the third largest in the past 9 months, during which there have also been three months where orders declined -4% or more.

Stepping back and taking a longer term look shows that core capital goods orders (black in the graph below) have been generally flat for the past two years, while total orders (blue) may have risen and then fallen a little. This is similar to the trajectory of manufacturing production (red) which peaked in late 2022, but has only declined about -1% since then:

The YoY look at the same data shows just how “unchanged” the trend has been:

When we compare with the 25+ years before the pandemic, we see a number of instances - 1998, 2012, 2015-16, and 2019 - where both new orders and production declined significantly into negative territory YoY without a recession occurring:

The sideways trend is also apparent in manufacturing employment (blue in the graph below), which has stayed in a 0.2% range for the past 18 months. Average weekly hours (red) has declined -1 hour or more, which before the China shock of 2000 and since had always meant a recession. But hours above 40.5 per week are mainly about overtime; thus since then the decline must go below 40.5 hours to be signficant:

Finally, turning from the production to consumption side, real personal consumption on goods, which was similarly rangebound from 2021 through the first half of 2023, has been on an increasing trend since:

While having both housing and manufacturing in a rangebound, mainly flat trend isn’t good, it isn’t recessionary either, as services provision and spending continue to perform very well.

The range-bound new home sales market continues

 

 - by New Deal democrat

As per my usual caveat, while new home sales are the most leading of the housing construction metrics, they are noisy and heavily revised. 

That was true again this month, as sales (blue in the graph below) increased almost 9% m/m to 693,000 annualized, after February was revised downward by -25,000 to 637,000. As the five year graph below shows, after the initial Boom powered by 3% mortgage rates, sales declined almost 50% in 2022, but have stabilized in the 650,000 +/-50,000 range for the past 16 months. For comparison I also include the much less noisy, but slightly less leading single family housing permits (red), which as anticipated appear to have started to follow sales down from their peak:

Here is a re-run of the graph I posted last week, showing the differing trajectories of new vs. existing home sales, showing that existing home prices remained elevated longer, and have taken longer to decline, by 40% vs. 50%:

Because mortgage rates have risen somewhat in the past few months (from 6.67% to 7.10%, I expect this range in new home sales to continue, with a slight downward bias in the immediate months ahead.

Also unlike existing home sales, where inventory is being constrained by would-be sellers trapped in 3% mortgages and thus prices remain near all-time highs, the median price of new homes declined as much as -16% from peak at their lows last year, and are still down -13.3%:

But, like sales, on a YoY basis prices have stabilized, and are only down -1.9%.

As I almost always point out, sales lead prices. Thus as shown above the range-bound sales for the last 16 months are leading to more stable prices.

The bottom line is that I expect this range-bound behavior in sales and prices, as well as the bifurcation between the new and existing home markets to continue until such time as the Fed moves significantly on interest rates.

Real median wage and income growth through March continued the recent increasing trend

 

 - by New Deal democrat

This is an update of some information I last posted several months ago.

Real median household income is one of the best measures of average Americans’ well-being, but the official measure is only reported once a year, in September of the following year.

So right now the most recent official measure is for calendar year 2022 (when you might remember gas prices surged to $5/gallon). In other words, it’s hopelessly out of date.

There are several ways of approximating real median household income on a more timely basis available in the public data. 

For this purpose, wages are a very imperfect proxy, because income includes things like stimulus payments or debt relief during the pandemic, and also because - especially during the pandemic - layoffs were concentrated among low wage workers, thus distorting the averages higher.

The best proxies make use of personal income. We can also get information from total payrolls. The below graph shows both real personal income (blue) and real aggregate payrolls (red), both divided by population. Here’s the data starting before the pandemic:

And here is the close-up after the end of pandemic related stimulus payments:

The big difference between the two is that real payrolls only include wages and salaries, while real incomes includes all sources of income, including stimulus payments and things like social security. Thus real per capita payrolls declined sharply during the fist months of the pandemic, and did not recover until late 2022, while incomes soared due to the pandemic related programs. Further, real payrolls stalled during 2022, while real incomes per capita actually declined.

Since late 2022, both measures have consistently increased. 

Additionally, a few private services have been able to use monthly data from the survey that gives rise to the jobs report to create a far more timely and illuminating monthly update. The best of these that I know of is Motio Research.

Here’s their most recent update, through March:

 

Like personal income, household income really spiked with the pandemic relief programs in 2020. It then went nowhere for almost three years, stuck at the same level it had been in 2019. Again, like personal income, that’s because of the spike in inflation, and the fact that jobs and real payrolls didn’t return to their pre-pandemic levels until 2022 and 2023 respectively.

The one remaining puzzle is why real *median* household income declined again into mid-2023, vs. *average* personal income, which increased.  One explanation might be the expiration of pandemic stimulus and relief programs, although I would expect that to show up in the broader income measure.

Some light can be shed by looking at *median* wage growth, as documented by the Fed:

Note that, compared with inflation, median (rather than average) wages continued to decline until early 2023. 

Another important explanation is likely that income growth has been concentrated among the the lowest quintile of households. In connection with the latest annual update, US News and World Report wrote:

 

While overall household wealth in America fell from the end of 2021 through the first three quarters of 2022, the bottom 20% of households by income saw their wealth grow.

“In total, household wealth for the lowest-income quintile rose by nearly 10% while wealth in all other income quintiles fell, according to figures from the Federal Reserve and nonpartisan data center USAFacts.

Here is the accompanying graph:

This very much helps explain why Biden’s approval ratings have been so poor throughout 2022 and 2023.

But, to return to the Motio Research graph, note that since last June, the trend has been rising again, and in March real median household income reached its highest level ever except for the 2020 stimulus months. What this means is that, if real household income growth had been concentrated among the lowest quintile through 2022, by mid-2023 it had spread upward to include the median group as well, and with some fits and starts this growth has continued.

Which is good news for the average American household.

Weekly Indicators for April 15 - 19 at Seeking Alpha

 

 - by New Deal democrat

I neglected to pt this up yesterday, so here it is now. My “Weekly Indicators” post is up at Seeking Alpha.

There continues to be a fair amount of churn and noise in the short leading and coincident time range. Nevertheless, the underlying theme is one of positivity. Aside from the swoon in the stock market this past week, the other big move was in industrial commodities, which spike higher late in the week. This is the first time they have been positive YoY in well over a year.

Typically that is because of higher demand straining against current supply, which means an expanding economy (with inflationary pressure building up).

As usual, clicking over and reading will bring you up to the virtual moment on all of these trends, and reward me with a little pocket change for organizing the data and bringing it to you.

The bifurcation of the new vs. existing home markets continues

 

 - by New Deal democrat

The bifurcation of the new vs. existing home markets continued in March, per the report on existing home sales and prices yesterday. Remember that, unlike existing homeowners, house builders can vary square footage, amenities, lot sizes, and offer price and/or mortgage incentives to counteract the effect of interest rate hikes.

On a seasonally adjusted basis, existing home sales declined from 438,000 to 419,000 in March. But this is well within the seasonally adjusted range of the past 16 months (gray, right scale in the graph below){also, note I am using Trading Economics graphs due to restrictions put on FRED by the Realtors; also note difference in scales):

At their worst seasonally adjusted levels last year, existing home sales were down over 40% from their 2021 peak. Meanwhile new home sales (blue, left scale), at their low in July 2022 down almost 50% from their 2021 peak, responded to mortgage rates by rebounding during much of last year before fading again in the past few months. They are presently down 35% from peak.

Some of the difference in trajectories between new and existing home sales can be explained by prices. Because so many homeowners have been frozen in place by their existing 3% mortgages, the inventory of existing homes for sale remains low, and that has driven prices higher almost consistently since the onset of the pandemic:

Although I can’t show you a graph, similar to the trajectory of the FHFA and Case Shiller repeat sales indexes, the median price for existing homes briefly turned negative in early 2023, troughing at -3.0% YoY in May. Thereafter YoY comparisons increased to a peak of higher by 5.7% in February. In March median prices were higher by 4.8%, which may or may not just be a pause.

Meanwhile the median price for new houses was down by -7.6% YoY in February. The below graph shows actual median prices for the last 5 years of new homes vs. the last 12 months (all that Realtor.com allows FRED to publish) of existing homes:

Median existing home prices are currently about 40% higher than their immediate pre-pandemic level, while new home prices are about 30% higher. 

Ultimately both the new and existing home markets are driven by mortgage rates. With a diminished supply of existing homes (because of prospective sellers being frozen in place by their current mortgage rates), their relative scarcity has driven prices comparatively higher than for new homes, especially as builders have moved aggressively to bring the purchase price of new homes down. This bifurcation will continue until the Fed moves significantly on rates.

Initial jobless claimZzzzzzzzzz . . . .

 - by New Deal democrat

For the last 8 months, initial and continuing claims have been remarkably consistent. Initial claims have varied between 194,000 and 228,000, and continuing claims have with the exception of three weeks right at the new year varied between 1.787 million and 1.829 million.

That rangebound trend continued this week as initial claims were unchanged at 212,000, and the four week average was also unchanged at 214,500. With the usual one week delay, continuing claims rose 2,000 t0 1.812 million:

Indeed, with the exception of last spring, initial claims have been essentially rangebound for the entire last 2 years!

For forecasting purposes, the YoY% change is more important. There, initial claims are down -5.5%, the four week average down -3.8%, and continuing claims are higher by 4.3% — still the lowest YoY reading for continuing claims in the past 13 months:

Needless to say, this suggests continued economic expansion in the next few months.

A reader over at Seeking Alpha several weeks ago asked what these looked like compared with population, since that is a more true measure of the tightness of the jobs market. Here’s the post-pandemic look:

The 4 week average of initial claims is 0.13% of the entire civilian labor force, while continuing claims are 1.1%.

Let’s compare that with the entire pre-pandemic record:

The 4 week average of initial claims is tied with the lowest ever pre-pandemic reading it had in 2019, while continuing claims are lower than the entire 50+ year pre-pandemic period except for 2017-19.

This in short remains a very tight labor market, where finding a new job is easier than at almost any time ever before the pandemic.

Finally, let’s update the Sahm rule implications with the first two weeks of April under our belt. Remember that both initial and continuing claims lead the unemployment rate, the former by more than the latter:

As per form, the unemployment rate followed jobless claims higher last year. Initial claims are now lower again, and continuing cliams remain flat. This suggests no further upward pressure on the unemployment rate in the months ahead, and likely some downward pressure towards 3.7% or 3.6%. In short, the Sahm rule is not going to be triggered.

Industrial production for March is positive, but the overall trend remains flat

 

 - by New Deal democrat

Industrial production, one of the premier series the NBER has historically used to declare recessions vs. expansions, has faded in importance since China was admitted to regular trading status in 1999. As you can see in the first graph below, both total and manufacturing production peaked in 2007. Further, manufacturing has continued to fade, as its post-pandemic peak has not equaled its 2010’s peak either:

In March, total production increased 0.4% from an upwardly revised, by 0.2%, February; but it is still down -0.6% from its September 2022 post-pandemic peak. Manufacturing production increased 0.5%, but is also down, by -0.2% from its post-pandemic peak as well:

Before the “China shock,” a YoY downturn in industrial production almost always meant recession. As the YoY graph below shows, there was a significant “industrial recession” in 2015-16 without any generalized economic downturn:

Whether the 2019 downturn would have resulted in a recession by itself had the pandemic not intervened will always remain an unanswered question. But again in 2023 production was again down YoY with no recession. As of March, manufacturing production is flat YoY, while total production is now up by 1.0%.

Bottom line: while March was positive, the overall trend remains generally flat.

Simultaneous declines in housing permits, starts, and units under construction in March suggests seasonality glitch, not a change in trend

 

 - by New Deal democrat

There was a big decline in housing starts last month, and a smaller but significant decline in permits. Whether that signifies a change in trend or just noise is the issue. I lean towards the latter. To wit, in reaction to both January and Feburary’s housing construction report I wrote, “To signify a likely recession, units under construction would have to decline at least -10%, and needless to say, we’re not there. With permits having increased off their bottom, I am not expecting such a 10% decline in construction to materialize.” I also indicated that I expected to see more of a decline in the actual hard-data metric of housing units under construction.

That is still the case.

To recapitulate my overall framework: mortgage rates lead permits, which lead starts, which lead housing units under construction, all of which lead prices. Of those metrics, the least noisy one that conveys the most signal vs. noise is single family permits.

In response to inflation data which generally stopped declining towards the holy 2%, mortgage rates have risen about .25% since the end of last year. For March as a whole, they averaged 6.82%. This is about average for the past 18 months, in which overall they have varied between 6.1% and 7.8%. In response permits have stabilized in the range of 1.42 million to 1.52 million units annualized. In March they declined -65,000 to 1.458 million annualized:

The relationship shows up even better when we compare the two series YoY:

With mortgage rates higher by a slight 0.25% YoY, permits went slightly positive YoY and are still higher by 1.5%.

As per usual, starts (light blue in the graph below) are the noisier of the metrics, declining 228,000 to 1.321 million annualized in March. Permits (dark blue) declined -65,000 to 1.458 million, and single family permits (red, right scale) declined -59,000 to 973,000:

These are among the poorest numbers for each in the past 12 months, but the simultaneity of the downturn (as opposed to a 1-2 month lag in starts) makes me suspect there may be a seasonal adjustment issue in play, perhaps having to do with Easter. Still, there isn’t enough there to break out of their range, and as discussed above mortgage rates have not suggested one is coming.

Next, to reiterate: housing units under construction (red in the graph below) are the best measure of the actual economic activity in the housing market. Here’s the long term historical view:

Those also declined, by -15,000, to 1.646 million units annualized:

Once again note the synchronicity of the downturn, making me suspect a seasonality glitch. Further, they are only down -3.7% from their peak, nowhere near the historical -10% most consistent with the onset of a recession.

Below I have broken out single vs. multi-family construction. Because, in response to record high house prices, builders turned to higher density, lower cost apartment and condo construction. Hence the record high last year in that metric. Last month multi-family construction faded slightly, while single family units under construction actually continued their slightly increasing trend:

As I wrote last month, I do expect a further gradual decline in total housing units under construction in the months ahead, to catch up with the decline in permits that bottomed one year ago. Here’s the post-pandemic view of starts, permits, and total units under construction:

But, as shown above, I doubt we will cross the -10% threshold that it would normally take to signal a recession, given the general stabilization of both permits and starts over the past 16 months.

Real retail sales rebound, forecast a continued “soft landing” for jobs growth

 

 - by New Deal democrat

As per usual, real retail sales is one of my favorite indicators, because it gives so much information about the consumer, and since consumption leads employment, it helps forecast the trend in the latter as well.

And the news this morning was good, as nominally retail sales increased 0.7% in March, while February’s number was revised higher by 0.3% to 0.9%. After accounting for 0.4% inflation in March, real retail sales increased 0.3%, and February was revised up to 0.5%.

To the extent there was bad news, it was that January’s -1.2% decline has still not been completely erased.

To the graphs: first, below I show the historical record for the past 15+ years of both real retail sales (dark blue) and real personal spending on goods (light blue), a similar but more comprehensive measure. The two metrics tend to trend together over time, although the latter has tended to increase more (hence I adjust to bring the trends more in line):

Here is the close-up post-pandemic view:

Real retail sales are still -2.9% below their April 2022 peak, and also about -1% below their nearer term August 2023 peak. Real spending on goods has been more positive. More importantly for the long term trend, real spending on goods has now completely caught up with real retail sales. The bigger picture is that real retail sales have trended neutral, while real spending on goods has trended higher.

Turning to the effect on employment, here is the longer term YoY% gains in both spending measures /2, which is the best match to forecast the near term trend in jobs (red):

Employment doesn’t respond to every noisy move in spending, but does tend to peak and trough about 6 months after spending, and responds to the longer term trend. If FRED allowed 6 or 12 month moving averages, the correspondence would be much closer.

With that caveat, here is the post-pandemic close up:

Historically negative YoY comparisons in real retail sales have usually meant recession, while positive comparisons have almost always meant continued expansion. Needless to say, that didn’t happen in 2022-23. The overall trend since mid year 2023 has been “less negative” to neutral, while real spending on goods has remained positive.

And as those YoY comparisons in consumption have improved, we have seen the decelerating trend in employment shift to a more consistent “soft landing” scenario. Thus real retail sales are forecasting continued growth in the neighborhood of the last few months’ numbers going forward through most of this year.