The bond market is sending a message

 

 - by New Deal democrat

The bond market is sending the T—-p Administration, and the US government as a whole, a message. It believes the US is in a secular inflationary trend. 

This is not only because T—-p’s tariffs, and his boneheaded war with Iran, are inflationary, but also because last year’s Big Budget Bust-out Bill which doles out $trillions to billionaires, the attempt this year to vote more $Billions to ICE, and even T—-p’s personal raid on the Treasury yesterday, all are gradually moving the US closer towards an unsustainable fiscal situation. 

To wit, yesterday yields on the 30 year US Treasury bond (dark blue) made a new nearly-20 year high, at 5.14%. The 10 year Treasury (light blue) also broke out to the upside, with a yield of 4.59%, which is not quite at the peak of its post-pandemic range, but close. Even the 2 year Treasury (orange) made a new one year high. The below graph also shows the 3 month Treasury (gold) and the Fed funds rate (red), all over the past four years beginning with when the Fed started to raise rates:

What is noteworthy over this time scale is that, with the exception of the 3 month duration, none of the longer maturities declined in yield over the past nine month, even as the Fed lowered interest rates.

And since the start of the war with Iran, all of the longer maturities have increased in yield by about .5%, as shown in the below close-up of this year:

The same dynamic has caused mortgage interest rates to rise to a new nine month high as well, which will not help the already recessionary housing market:

The emerging economic dynamic, to paraphrase Mark Twain, won’t repeat the inflationary 1960s-70s, but it likely will rhyme. Showing how that is likely is a little difficult with graphs, but bear with me.

Here are two graphs, of the 1960s and 1970s, showing (1) the increase, in percent, of the 10 year US Treasury (*10 for scale), in blue. If in a given quarter the average yield increased 0.5%, that is shown as +5% in the graph. Also in the graphs are the nominal (gold) and real (red) quarterly percentage change in GDP:

Here is what I want to direct your attention to. Inflation and the secular increase in interest rates began to take off in the mid-1960s with LBJ’s “guns and butter” stimulative economic policy. Thereafter, as interest rates and inflation both increased in any given quarter, in real terms growth in GDP was likely to decelerate or even decline. Thus, for example, in the fist part of the 1960s real GDP growth averaged about 2% annualized, but after the middle of the decade tailed off to about 1% annualized with the exception of the first two quarters of 1968. Similarly, in the 1970s real GDP growth approached 0% in most quarters where there was a significant increase in bond yields.

I expect a similar secular dynamic is beginning to take hold now. And it will likely last throughout the next decade or so. On a long term basis, bond yields tend to follow a cycle which lasts approximately one average human lifetime (i.e., polities are most likely to forget the historical lessons and repeat the same historical mistakes that they have not lived through). In the case of the US, there was a cycle that began with an interest rate peak coincident with with Civil War and lasted until 1920:

Then there was a second cycle that began in 1920 and ended in about 1981:

And we are currently living through the next cycle, that began in 1982 and made its trough in the last decade:

As I’ve noted elsewhere, the Founders’ generation, like the political theorists of earlier times, worried that in a democracy the masses would likely simply vote themselves money out of the Treasury. What they did not foresee was that the very wealthy would put together a victorious coalition and then vote *them*-selves money out of the Treasury.

I cannot say what will happen with bond yields in the next day, month, or year. But the evidence is clear and convincing that an inflationary secular trend of rising interest rates is here.

What’s driving the stock market Boom (or Bubble)?

 - by New Deal democrat

In the past month or two, a number of times I have been asked by friends and neighbors why, if the economy feels so bad, does the stock market keep making new record highs? And is this a Boom or a Bubble?

Here is what I have been telling them.

There are two current drivers of the stock market: (1) AI data center related spending, and (2) domestic energy company profits.

Let me tackle the second one first, because it is relative simple and straightforward. As we all know, the Strait of Hormuz, through which most of Middle Eastern oil flows, remains closed. But the Gulf of Mexico is wide open! And that means that countries and companies that need oil are bringing their empty tankers to Gulf Coat ports and loading up. US oil exports have thus hit a new record:

Further, because the price of oil is set globally, they are loading up at that price. Which means that domestic producers like ExxonMobil and Chevron are making out like bandits:

So while consumers may be suffering at the gas pumps, US based producers most certainly are not.

Now let’s turn to AI data center related spending. Below are three components of industrial production that are tied to AI data centers: semiconductors (blue), computer and electronic products (orange), and electric and gas utilities (gold, right scale):

All three have increased much more than industrial production as a whole over the past several years, particularly semiconductor production, which is up over 60%! As I have written before, once you take out AI related spending, manufacturing and production have gone basically nowhere in the past several years.

So, in answer to the first question, these two sectors are what have been driving the stock market higher.

But is it a (sustainable) Boom, or an (unsustainable) Bubble?

To give you my sense of that, I have to get into the weeds on some data I don’t normally concern myself with, which are called “stock market internals.”

Last week I wrote that half of all S&P 500 profits have come from just 5 companies:

That’s not exactly a widespread Boom.

Next, here is an update of a stock market indicator I touched on a few times last year: the advance-decline line. This tells us how many more stocks are increasing in value than decreasing. If the economy is doing well, such increases ought to be widespread. But if only a few companies are advancing, that tells us that the majority of the economy is not doing so well. Here’s what that looked like late last year when I wrote about it:

And here is what it has looked like so far this year:

While compared with last year, there has been an increase in the total number of advances vs. declines, in the past month this number has fallen fairly sharply, back to where it was in February, and only a little better than late last year.

A similar lack of widespread health is shown by the number of new highs vs. new lows in stock prices:

In the recent advance, only a few more stocks made new highs than made new lows.

Finally, let me touch on one other measure of relative health: the stock vs. bond dividend gap. This is the difference between the amount of dividends paid on S&P 500 stocks, vs. the interest rate an investor can get by holding bonds. The higher the gap, the more investors are relying on stocks rising in prices to cover their risks.

And to start with, the S&P 500 is at or near record low dividend payments, at only 1.06% of their price:

Dividends on the S&P 500 have generally drifted lower for decades, but at least during 1981-2020, so did interest rates on bonds.

So here is how the current dividend yield on the S&P 500 compares with the yield on a 10 year Treasury, currently at about 4.5%:

And here is the same comparison with the 2 year Treasury note:

Both of these measures are at their worst levels for stocks since before the Great Recession 20 years ago.

Finally, here are the yields on AAA and BAA investment grade corporate bonds:

These are at levels equivalent to just after the Great Recession. Put another way, you are currently being paid about 3.5% more than S&P 500 stock dividends to own a 10 year Treasury, about 4.5% more to hold an AAA rated corporate bonds, and about 5.0% more to own a BAA investment grade bond. All to bet that the stock market will continue to increase in price in the next year. 

This does not tell us in *absolute* terms whether stocks or bonds are a “good” deal. What it does tell us is that in *relative* terms, you are taking more risk than at any point in the past 20 years to gamble on stock price appreciation making up for the puny dividends you are earning on them. Bonds could still be a bad deal at present — but stocks are relatively speaking at far more risk.

The present situation reminds me very much of 1999, when the advances in the stock market were focused on the dotcom issues. But the advance decline line deteriorated for about a year before the dotcom implosion kicked in. 

So my conclusion is that, at the moment, the momentum in the AI sector, joined by the windfall profits for domestic oil producers, are more than outweighing the difficulties by most consumers. But that momentum looks extremely risky. Like a bubble.

 

Weekly Indicators for May 11 - 15 at Seeking Alpha

 

 - by New Deal democrat

My “Weekly Indicators” post is up at Seeking Alpha.

Unsurprisingly, the big move among the high frequency indicators in the past week was interest rates. There was also a secondary big move, which is also inflationary, and reflects the ongoing idiocy of the current Administration in Washington: international shipping rates spiking, in one case to a 24 month high.

As usual, clicking over and reading will bring you up to the virtual moment as to the state of the economy, and toss me a penny or two towards my next outing for lunch.

Two manufacturing reports show increased expansion, but with a nasty side of inflation

 

 - by New Deal democrat

One of the things I used to harp on was that many forecasters make the mistake of simply taking an existing trend and projecting it forward. Often it is paired with the idea of “all things being equal,” i.e., that there won’t be countertrends in other sectors of the economy. This is one of the big reasons I have relentlessly updated my “weekly high frequency indicators” for about the last 15 years - because although they can be noisy, it is where you will first pick up a change in trend.

One prime recent example of this was when I noticed late last year that the regional Fed manufacturing indicators were improving - despite the “Liberation Day” tariffs and the general chaos coming out of Washington. It appeared that manufacturers had found a modus vivendi and had adapted to the new environment.

And that is what we continued to see this morning, both with the Empire State Manufacturing Survey and the Industrial Production report.

I used to call Industrial Production “the King of Coincident Indicators,” but with the shrinking of manufacturing as a share of the US economy to about 25%, that is no longer the case. Nevertheless, it is an important coincident indicator, with the emphasis on *coincident.* It doesn’t tell us where we are going, but is an important signpost about where we *are.*

For the past six months or so, the general trend in manufacturing production had been rangebound, contrasted with a sharp increase of likely AI-related spending in utilities. There was an important change this month, as manufacturing (red in the graph below) broke out of that range to the upside, increasing 0.6% for the month, while the more noisy utility production (gold, right scale) increased 1.9%, driving total production (blue) to a 0.7% increase and a new post-pandemic high:

Similarly, both the coincident headline number (blue) in the Empire State Manufacturing Survey and the leading new orders component (red) increased to new 4+ year highs at 19.6 and 22.7, respectively:

Just one month and just one regional survey, but the new orders subindex has been positive every month but one since last October, and has broken out to the upside in the last two months.

Employment in this regional survey (blue) has also been positive for the past few months. The big problem - and this has extended to virtually all the regional surveys, both in manufacturing and services - is on the inflation side. Both increases in prices received (orange) and even moreso prices paid (red) have turned extremely widespread, with new highs this month not seen since the summer of 2022, and close to their post-pandemic highs:

So, the message from this morning’s reports is mixed: manufacturing has been picking up, but the already bad inflation appears to be worsening.

April retail sales: consumers may be switching to “wait and see” mode

 

 - by New Deal democrat

Consumer spending is about 70% of the economy, and retail sales is our first wide measure of that spending. For the second month in a row, this morning’s update for April was unsurprisingly dominated by what happened at gas stations. 

Nominally, total retail sales rose 0.5% in April, But after taking the monthly 0.6% increase in consumer prices into account, real sales declined -0.1% (blue):

But let’s take a look at the impact of gas prices. The below graph shows total retail sales (blue), and also retail sales minus gas (gold), both nominally. As per the above, total sales rose 0.5%. But sales ex-gas only rose 0.3%:

Now let’s look at real sales ex-gasoline. The below graph shows real sales ex-gasoline, deflated two ways; first, by the headline CPI number (blue); and second, by CPI excluding energy (gold):

Deflated by headline CPI, retail sales excluding gasoline declined -0.4% in April, and remain below their peak last August. Deflated by CPI less energy, they declined -0.1% from their peak in March. And although I won’t show the graph this month, it is interesting that the same phenomenon occurred in the 20006-07 period before the Great Recession, where total real retail sales trended slightly higher, but excluding gasoline trended lower.

On a YoY basis, nominal total retail sales were up 4.9%, but in real terms were only up 1.1%:

Excluding gasoline, nominal sales YoY were up 3.7%, and in real terms *down* -0.1% using headline CPI, and up 0.8% using CPI excluding energy:

In the March personal income and spending report, we saw that consumes handled the first month of gas prices increases by essentially just putting the extra spending on their credit cards. While it is certainly within the range of noise, the retail sales report for April suggests that consumers have transitioned to a “wait and see” pattern, withholding additional purchases until they see what happens next with gas prices.

Finally, since consumption leads employment, here is the update of YoY total real sales (/2 for scale) together with employment (red):

This suggests that on a YoY basis the stabilization we have seen in the last two jobs reports is likely to continue in the next several months. Which means it is not likely to be recessionary, but more likely to be in “wait and see” mode as well.

How mass immigration raids might explain the historic lows in jobless claims

 

 - by New Deal democrat

I want to write an extended discussion of the April retail sales report released this morning, to tease out the effect of the spike in gas prices, but first let me do my usual weekly update on jobless claims. These have historically been a good short leading indicator for the economy, but it is possible that is being distorted now.

As per the last six months, new jobless claims, along with the Boom (or maybe bubble) in the stock market, are the most positive of all of the current economic indicators. For the week, initial claims rose 12,000 to 211,000, while the four week moving average rose 750 to 203,750. Although I won’t put up the long term historical graph, these continue to be lower than at any point between 1970 and 2018, and at the low end thereafter. Continuing claims, with the typical one week delay, rose 24,000 to 1.782 million. Historically this is also lower than at any point between 1975 and 2018, although they were much lower in the immediate post-pandemic Boom of 2021 through mid-2023:

As per usual, the YoY% change is more important for forecasting purposes, and so measured initial claims were down -6.6%, the four week average down -11.1%, and continuing claims down -5.4%:

Since it is early in the month, I’ll dispense with my typical look at what that might mean for the unemployment rate. But let me briefly address why jobless claims might be so historically low, particularly as the US population has doubled since 1970. Indeed, as a percent of the labor force, or of the US population as a whole, the number of claims in the past six months has been tied for the lowest ever with brief periods of time in 2022 and the end of 2023, as shown in the below historical pre-pandemic graph that is adjusted to show the current rate at the “0” line:

Is business humming in a way it never has before? Probably not. The most likely explanation has to do with immigration and the deportation raids that became more widespread in the past year. 

Let me state right off the bat there is really no good, on-point statistic that I can find. The situation is opaque, but if we look at monthly employment and unemployment statistics by native vs. foreign born status, and by ethnicity, at least some light may be shed.

Every month the Household Survey in the Employment report includes numbers for the employment level of native vs. foreign born persons. Since these are not seasonally adjusted, the best way to look at them is the YoY% change:

I should note that the Household Survey has shown absolute YoY declines in employment for the past several months. What is noteworthy for this discussion in the above graph is that the YoY employment level for immigrants turned negative rapidly last summer as immigration raids became more widespread, and with one exception has remained negative ever since.

Similarly, the Household Survey tracks the unemployment rate by several ethnic groups, including Hispanic or Latino. Again, these are not seasonally adjusted, so the below graph tracks the change in the percent unemployed YoY:

Note that the change in the unemployment rate turned lower for immigrants (gold) by last summer, and with two exceptions has remained lower ever since, even as the unemployment rate in total (dark blue) and for the native born (purple) has remained higher. And the unemployment rate for Hispanics has generally been the lowest of all (red).

Either uniquely immigrants, especially HIspanics, are not being laid off — *extremely* unlikely — or they are not making claims or even willing to discuss their job status with Census Bureau phone interviewers, which is much more likely. This seems particularly true since both the employment level and the rate of those who are unemployed have *both* declined, suggesting a reluctance to interact with the system.

I want to point out that the above explanation is necessarily speculative, and based on information that remains murky. But it does seem to explain the unusual information about jobless claims we’ve been getting for this year.