Resurrecting the metric: initial claims lead the unemployment rate; no recession signal so far

 - by New Deal democrat

 Initial jobless claims declined -2,000 to 229,000 last week, vs. the 50+ year low of 166,000 set in March. The 4 week average rose 4,500 to 223,500, compared with the all-time low of 170,500 eleven weeks ago.  Continuing claims rose 5,000 to 1,315,000, which is 9,000 above their 50 year low of 3 weeks ago:

Initial claims have been in an uptrend over the past 2.5 months. If this continues until the end of this month, they will no longer qualify as a “positive” in my array of short leading indicators, although they have not risen to levels that would change their rating to a negative.

Since the normal DOOOMers are baying that we are already in a recession, now is a good time to resurrect the construct that initial jobless claims lead the unemployment rate.

Why? Well, for example, the Sahm Rule is that when the 3 month moving average of the unemployment rate rises by 0.5% relative to its low in the previous 12 months, you’re in a recession. That’s somewhat conservative. On at least two occasions, 1953 and 1970, the unemployment rate only went up 0.1% for 1 month before a recession started. Paul Volcker started a recession in 1981 where the unemployment rate hadn’t moved up at all!

But in general, while the unemployment rate is a lagging indicator coming out of a recession, it is actually a negatively over-sensitive one, as it is a  slightly *leading* one going into recession.

With that in mind, here is the long term graph of initial jobless claims (red) vs. the unemployment rate (blue, right scale):

Typically a uptrend in initial claims leads an uptrend in the unemployment rate by 2-4 months.

Here’s the past two years:

The very mild uptrend in initial claims we’ve had in the past several months is similar to the one at the beginning of 2021 during the first winter wave of the pandemic. That led to a pause in the decline of the unemployment rate a few months later during spring 2021.

At worst, the current uptrend in claims (so far!) is consistent with a potential 0.1% uptick in the unemployment rate going into autumn.

Which means, while it’s not impossible, it’s very unlikely that the unemployment rate will signal the onset of a recession during that time.

Housing unaffordability closes in on bubble peaks; expect substantial price declines and increased foreclosures in the likely oncoming recession

 

 - by New Deal democrat

I last looked at the issue of housing affordability at the beginning of April. As we all know, mortgage rates have continued to skyrocket in the past several months. At present they are just under 6.10%:

This has changed the calculus on housing affordability considerably. So let’s take a look.

In April house prices in real terms were already almost identical to their 2006 highs. Depending on what house price index you use, nominally prices are up somewhere on the order of 60% since then. For example, here is a graph of new home prices and the FHFA index for the 2000s:

And here is the past several years:

Meanwhile the median price for an existing home peaked in July 2006 at $230,200. As we saw yesterday, as of last month they were $402,000. 

Average hourly wages for non-supervisory workers are also up a little over 60% since 2006:

Since in real, wage-adjusted terms, house prices are about the same now as they were at the peak of the housing bubble, let’s compare an identical mortgage then and now as well, for simplicity’s sake using $250,000, at the prevailing mortgage rates. Here’s the monthly payment for each:

April 2006: $1865.

July 2006: $1913.

April 2022: $1583.

June 2022: $1798.

The bottom line is that the average monthly mortgage payment has increased by about 13.6% in the past two months, and now is about 96% in real, wage-adjusted terms, of what it was at the peak of the bubble.

Even before yesterday’s report, the NAR’s “Housing Affordability Index” had dropped to 105 in April. How low is that? Well, for comparison here’s what the Affordability Index was during the 2000s:

At the worst of the housing bubble, it sat right at 100.

In short, housing is equal to its worst affordability levels in the past 35 years.

Frankly, I have been very surprised at this, since mortgage lenders are being much more careful now than they were at the peak of the bubble, when anything that fogged the mirror could get a loan.

This changes the calculus of the housing market - and the economy - going forward significantly. 

While I don’t see the banking system fallout that we had in 2008, when all the birds came home to roost on those ridiculous loans, creating a cascade of financial system defaults, I *do* now see it as being almost inevitable that there will be another significant decline in house prices that will last a number of years. This will trap a large number of younger homeowners in houses that are financially “underwater,” as they make payments on a house they can’t sell for the price at which they bought. And an increase in unemployment during a recession next year that looks increasingly inevitable will mean a substantial increase in foreclosures for some of these buyers as well.

Not so good.

Existing home sales: the freight train of price appreciation rolls on

 

 - by New Deal democrat

Although existing home sales are less economically important than new home sales, what has been happening with their prices, given the experience of the housing bubble and bust 15 years ago, is of added importance.

The simple summary is that sales have declined substantially, while price appreciation keeps rolling on.

Sales of existing homes were down 3.4% for the month, seasonally adjusted; down 8.6% YoY; down almost 20% from their January 2021 peak; and, at 5.41 million annualized, the lowest level since June 2020:

This isn’t a crash - at least, not yet. But it is certainly at a level consistent with an oncoming recession.

The story is completely different as to prices. At $407,600, the median price of an existing home increased 4.8% for the month and 14.8% YoY (Note: prices aren’t seasonally adjusted, so the YoY view is the best measure; graph does not show this morning’s data):

Prices were up over 25% YoY last June, so while this is a 40% deceleration, it is consistent with continued price appreciation if we were able to seasonally adjust.

Additionally, in May inventory was still down -4.1% YoY. Importantly, the NAR’s weekly update showed inventory increasing YoY in the last week of May and the first several weeks of June, so this may be the last hurrah for that metric.

Bottom line: Prices follow sales, and will in this case as well, but they haven’t - yet.

Coronavirus dashboard for June 19: documenting the transition from pandemic to endemic

 

 - by New Deal democrat

The COVID-19 pandemic is ever so gradually transforming into an endemic illness, the major risks of which still mainly fall on seniors.

Here is the long-term view of cases (dotted line) and deaths (solid line) in the US:

While cases are similar to the peaks of 2020, but far below those of 2021, deaths are lower than at any point except for June and July last summer.

Similarly, hospitalizations remain lower than at any point except several months during summer 2020 and 2021:

Focusing on the last 3 months, cases have been generally unchanged in the range of 100-110,000 for the past month, and deaths have varied between 250-330 for the past 1.5 months (the several gaps higher in deaths are due to periodic data dumps by North Carolina that can be ignored):

Essentially, both cases and deaths have plateaued at current levels, despite the changing landscape for variants, as Ba.2.12.1 replaced Ba.2, and now Ba.4&5 are increasing.

Here is the latest from Biobot, which tracks wastewater, an early warning system:

“Real” cases are down sharply in the Northeast, where Ba.2.12.1 was most prevalent, and slightly down in the Midwest and West, but rising again slightly in the Midwest.

We should know within a week or two whether the emergence of Ba.4&5 as the dominant strain is going to create any renewed wave or not. So far, very preliminarily, the answer is “not.”

Weekly Indicators for June 13 - 17 at Seeking Alpha

 

 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

My paradigm is: first the long leading indicators turn. Then the short leading indicators turn. Then the coincident indicators turn. Finally the lagging indicators turn.

For months I have been documenting the downturn among the long leading indicators. In the past few weeks, that deterioration has been gradually spreading among the short leading indicators.

As usual, clicking over and reading will bring you up to the virtual moment as tho the state of the economy, and will bring me a small reward for my efforts as well.

Positive production print points to continued economic expansion in May

 

 - by New Deal democrat

The usual suspects are out, claiming that a recession has either already started or is imminent. Well, the big reason I call industrial production the King of Coincident Indicators is because empirically is the one whose peaks and troughs coincide most definitively with NBER recession dates. And unless there is a significant downward revision, in May the King of Coincident Indicators proclaimed: no recession yet.

Total production rose 0.2%, while manufacturing declined -0.1%. April’s overall number was also revised higher, from 1.1% to 1.4%, while manufacturing remained at +0.8%. The former made yet another new record high:

On a YoY basis, total production is up 5.8%, while manufacturing is up 4.9%. Compared with the last 40 years, and particularly the last 20, this remains pretty good growth:

A recession *could* start from these YoY numbers (see 1990 and 2007), but usually YoY production is decelerating pretty rapidly before a recession actually begins.

A close-up of the monthly changes since the depth of the pandemic recession shows that May was weak, like much other data for the month, but not indicative of any significant trend change yet:

With an employment gain of nearly 400,000, and (again, unless revised lower next month) a positive print on production, overall the economy continued to move forward in May.

An across the board downturn for housing permits and starts in May

 

 - by New Deal democrat

Housing permits and starts declined across the board in May.

In the past year there has been a unique divergence between permits and starts due to construction supply shortages.  This has been reflected in the number of housing units authorized but not started increasing to a near-50 year records of 298.4 in March. In May that number increased from April by 1.5 million annualized to 287.6:

As a result, for the time being I am paying the most attention to the three month average of housing starts (blue in the graph below), as these reflect actual economic activity, vs. permits (gold) which are only potential activity. That three month average declined to 1.691 annualized, a 5 month low:

For the month, single family permits (red above, right scale) declined 61,000 annualized to 1.048 million, a 22 month low. Total permits declined 128,000 to 1.695 million annualized, an 8 month low, and starts declined for the month by 261,000 to 1.549 annualized, a 13 month low.

Finally, below is the most recent version of a graph I have run many times in the past 10 years, showing that mortgage interest rates (red, inverted *10 for scale) lead housing permits (gold) and starts (blue):

Interest rates are higher by 2% vs. one year ago. As the graph shows, the last time the comparison was this bad was 1994, resulting in a 20% decrease in permits and starts the following year. Both permits and starts have now turned negative YoY. 

The “demographic tailwind” that buoyed housing activity 5 and 10 years ago has dissipated, as the number of 25-35 year old first time buyers has stopped increasing. Thus I expect a 20% YoY decline in housing permits and starts to manifest over the coming 12 months. 

The conundrum is whether the 50 year high backlog in units not yet started will delay the downturn until it clears - which might take another 6 to 12 months. Since starts are the actual economic activity, until I see an unequivocal downturn there, the massive negative signal from permits, mortgage rates, and mortgage applications remains open to question.

The increasing trend in new jobless claims continues

 

 - by New Deal democrat

Initial jobless claims declined -3,000 to 229,000 last week, vs. the 50+ year low of 166,000 set in March. The 4 week average also rose 2,750 to 218,500, compared with the all-time low of 170,500 ten weeks ago.  Continuing claims rose 3,000 to 1,312,000, or 6,000 above their 50 year low of 2 weeks ago:

It’s now clear that initial claims have been in an uptrend over the past 2.5 months. If this continues until the end of this month, they will no longer qualify as a “positive” in my array of short leading indicators, although they have not risen to levels that would change their rating to a negative.

This is yet more slight weakening in the economic indicators, and yet more reason for concern about a recession as we get to 2023.

Negative May and YoY real retail sales add to the foreboding signals of a recession next year

 

 - by New Deal democrat

Nominal retail sales for the month of May declined -0.3%, and April was revised down by -0.2% to +0.7%. This reduces April’s number, after inflation to +0.4%, followed by a “real” decline in May of -1.2% after rounding. YoY real retail sales were up 8.1%, but because inflation in the past 12 months has been 8.5%, real retail sales YoY is down -0.4%. Here is a graph of the absolute value of real retail sales:

In the past 75 years, a decline in real retail sales YoY has frequently - but not always - indicated a recession. Here’s what the past 30 years look like:

Needless to say, not good news. 

Next let’s turn to employment, because real retail sales are also a good short leading indicator for jobs.

As I have written many times over the past 10+ years, real retail sales YoY/2 has a good record of leading jobs YoY with a lead time of about 3 to 6 months. That’s because demand for goods and services leads for the need to hire employees to fill that demand.  The exceptions have been right after the 2001 and 2008 recessions, when it took jobs longer to catch up, as shown in the graph below, averaged quarterly through the First Quarter: 

Now here is the monthly YoY comparison for the last year through May:

II have been writing for months that I have expected the blowout job numbers of about 500,000 per month to slow down to a range of about 100,000-300,000 per month by early autumn. The last 3 months have averaged 400,000, which has probably been the beginning of that slowdown.

Finally, real retail sales per capita is one of my long leading indicators. Here’s what it looks like for the past 25 years:

And here is the last year:

Last month’s good report temporarily switched the long leading signal from negative to neutral, but this month it goes back to negative - along with most of the rest of the long leading indicators, which have been increasingly foreboding about the economy next year.

On the road, and missing all the fun on Wall Street . . .

 

 - by New Deal democrat

I’m still on the road, and there is no important economic statistic to report, but I will make a brief market comment.

YoY stocks are now down 10%. Except for the very bad 1982 and 2008 recessions, and the 2000 Nasdaq bubble and the 1987 crash (which were prolonged bear markets), that has typically been close to their YoY lows:

And with rare exception that level has only been hit about midway through a recession.

The only time buying a broad basket of stocks at this price reduction has not been amply rewarded within after 2-3 years is during the 1929-32 Great Contraction.

< Sigh > I miss the days when I could read triumphalist diaries by DOOOMers at Daily Kos, who were infallible contrary indicators, and be sure that the bottom was in. 

On the other hand, here’s a 12 month graph of Tesla, recently valued at a level equal to if not exceeding the entire rest of the auto industry:

That’s a pretty severe price reduction. Until you zoom out to the 5 year perspective:

Is there any reason for Tesla still to be selling at 10x its price only 2 years ago?

And further, is there any reason for Bitcoin to exist at all?:

On the bright side, a lot of know-it-all smart mouthed 30 year olds are going to STFU.

Weekly Indicators for June 6 - 10 and a comment on COVID

 

 - by New Deal democrat

I’m still traveling, so light posting for a couple days more.

In fact, I neglected to post a link to my Weekly Indicators on Saturday. A bit tardy, here it is. Conditions across all time frames do not look so good, and with Friday’s poor inflation report, I expect the Fed to continue stomping on the brakes - especially since we know that the YoY change in rents and “owners’ equivalent rent,” which together make up about 1/3rd of all inflation, are likely to keep rising (since they lag house price indexes, which are still at or near their YoY high growth rates, by a year or more).

Meanwhile, there’s good news and bad news on the COVID front. For the last three weeks, cases have plateaued at roughly 105,000 per day, and deaths have varied between 275 and 340. So BA.4/5 does not seem to be creating any new “wave” at this point. Further, between April 4, the recent trough in cases, and one month ago, there were a cumulative total of about 2.150 million cases. With a one month lag, deaths totaled 12,300. That’s about a 0.6% fatality rate, or about 1 in 400 cases. So the good news is, over the population as a whole, COVID has recently had about a similar death rate to a bad year for seasonal flu.

The bad news is that the death rate is heavily concentrated towards the elderly. Among people under 50 years of age, in May there were only about 125 deaths total. For age 50-65, there were another 400; for 65-74, 700; for 75-84, 1000, and 85 and older, 1500. In other words, for seniors, and particularly those over age 80, COVID remains a deadly serious disease.

Consumer prices rise 1% in May alone; owners equivalent rent at 30 year high; expect the Fed to keep stomping on the brakes

 

 - by New Deal democrat

Today is a travel day for me, so I’ll keep this relatively brief.

People who were hoping inflation would abate did not get the news they wanted from the May CPI. Consumer prices rose 1.0% in that month alone. Inflation less energy rose 0.7%, and “core” inflation less food and energy rose 0.6%. On a YoY basis, prices are up 8.5%, tied for a multi-decade high with a few months ago. Core prices are up 6.0%, down slightly from their February and March peak:

The news was relatively “good” in the prices of used cars and trucks, up 1.8% for the month, but “only” up 1.6.1% YoY, down from 45% half a year ago:

Finally, as expected, house prices (as shown via the FHFA Index in blue below) have continued to bleed over into owner’s equivalent rent, which rose 0.6% for the month, and is up 5.0% YoY, a 30 year high:

Bottom line: the Fed is going to continue stomping on the brakes. The only issue will be whether we get an old-fashioned “bust” before short term rates get raised high enough to invert the yield curve, a la the late 1940s and early 1950s.

Initial jobless claims now in a clear uptrend, and other economic notes for the week

 

 - by New Deal democrat

First, a note on other economic news from earlier this week. 

The housing market is getting absolutely crushed by 10 year+ high mortgage rates. Mortgage applications fell to levels not seen since 2017 (except for a few weeks during the 2020 Covid lockdowns). Refinancing is at 20 year lows. The next questions are when prices will peak, and what will happen with the huge backlog in housing starts from properties already permitted? We’ll find out more in a week with the next report on housing starts and permits, but we won’t get more price information for a couple more weeks.

I saw a couple of comments on consumer credit increasing. As I’ve noted, the consumer is in a typical late expansion mode, where real income growth if flagging, and consumers are extending themselves on more credit. I don’t see anything drastic happening here.

I also read a few reports that retailers are reporting marked dips in sales in the past month. I think we have reached the point where gas prices are high enough that consumers are cutting back on other purchases. I doubt we’re at an “oil shock” point, though, where consumers *over*compensate becoming extra cautious about other purchases.

Finally, there’s been some discussion about the future course of inflation. One highly compensated Wall Street pundit claimed that consumer spending had not caused any inflation, because - wait for it - in inflation-adjusted terms, it had kept up with inflation. In other words, consumer spending behaved exactly as you would expect it to if more money were chasing the same amount of goods. Although that’s clearly *not* the case with gasoline and cars, where demand has never completely returned to pre-pandemic levels, and yet prices are much higher.

Meanwhile, Paul Krugman among others is touting that inflation may be subsiding, so the Fed should be wary of over-reacting. Measured by wages for non-managerial workers, it has abated just slightly, so I’m not so sure. Beyond that, as I’ve pointed out in a few recent missives, this is how Booms turned into Busts in the immediate aftermath of World War 2. The Fed never raised rates, but price and mortgage rate increases overwhelmed the ability of consumers to keep pace. So they didn’t. Inflation abated, because there were brief recessions which caused an abatement of interest rates and price increases. I.e., be careful what you wish for.

Turning to the news actually at hand  . . .

Initial jobless claims rose 27,000 to 229,000 last week, continuing above the recent 50+ year low of 166,000 set in March. The 4 week average also rose 8,000 to 215,000, compared with the all-time low of 170,500 nine weeks ago.  Meanwhile continuing claims were unchanged at 1,306,000, tied for a 50 year low:

It’s now safe to say that initial claims have been in an uptrend over the past 2.5 months. If this continues a few more weeks, they will no longer qualify as a “positive” in my array of short leading indicators. I will be interested to see if this is confirmed by a significant decline in the “job openings” category of the next release of the JOLTS report, which would indicate that employers are affirmatively cutting back on their hiring plans.

In any event, yet more (slight) weakening in the economic indicators, and yet more reason for concern as we get to 2023. A few weeks ago I went on “Recession Watch” beginning with Q1 2023 and noted that, because I rely in large part on Prof. Geoffrey Moore’s long leading indicators, which also were the basis for ECRI’s forecasts (the last time they were public about their metrics), I expected ECRI to start talking about recession as well.

Well, within the past week, via a Lakshman Achuthan interview on CNN, they have.

Coronavirus variant update: and on to the BA.4/5 wave

 

 - by New Deal democrat

Last week the CDC update showed variant “Ba.1.1.526” increasing quickly to 6.6% of all cases. Although they did not note it, I wrote that this was almost certainly Ba.4/5; they simply had not made the change yet.

Well, this week they did. This morning’s “nowcast” update of variants shows Ba.1.1.526 having vanished, with Ba.4/5 having taken its place, and having doubled to 13% of all cases in the past week, even as Ba.2.12.1 slowly increased from 59% to 62%:

Here’s the regional breakout:

Ba.2.12.1 now makes up a little over 80% of cases in NY, NJ, and PR, while BA.4/5 combined make up 18% of all cases in the northern Rockies and 22% of the southern Plains, including Texas.

It is unclear it this point how much, if at all, nationwide cases will increase. It depends on how much Ba.2.12.1 fades compared with how quickly Ba.4/5 take over. In South Africa, the Ba.4/5 wave rolled in and out very quickly earlier this spring:

There the Ba.4/5 wave rolled in over about a 3 week period, and rolled out almost as quickly. Cases were only 1/3rd as many as the original Omircron wave at peak, while deaths were only 1/5th as high. Hopefully the same will happen here. 

Coronavirus dashboard for June 6: transitioning ever so gradually into a less fatal endemic condition

 

 - by New Deal democrat

The economic calendar is very light this week, with no significant news until Thursday, so let’s take this opportunity to update the situation with COVID-19.

First, as of one week ago subvariant BA.2.12.1 continues to increase very slowly its share of overall cases, up to 59% nationwide:

This subvariant makes up 78% of cases in NY and NJ, but only 38% in the Northwest. Meanwhile new subvariants BA.4/5 (still listed as BA.1.1.529 by the CDC) has risen to 6.1% of all cases, with a high of 12.4% in the northern Great Plains, and a low of 2.9% in NY and NJ:

The BA.2.12.1 wave appears to have peaked 8 days ago, at 115,700 cases. It is presently down to 106,800. Meanwhile deaths have *still* continued to decline from their Omicron peak of February, making a new 10 month low of 258 on June 3rd:

Hospitalizations have continued to increase, up to just over 29,000, compared to their low of 10,264 on April 6:

The trend of COVID not being nearly so lethal as it was originally (regardless of the causation) continues. Two really important trends stand out.

The first is that we are now more than 2 full months after the post-Omicron trough, and yet deaths have continued to ever so slowly decline, as noted above making a new low just a few days ago. Typically, as is shown in the chart below, hospitalizations have made peaks and troughs simultaneously with or just slightly after cases, while deaths make equivalent peaks or troughs within 3-4 weeks later:

Event  Peak	Cases Date	Hosps Date	Deaths Date
Alpha	1/11/21	1/9/21	1/13/21
Delta	9/5/21	8/27/21	9/21/21
Omicron	1/15/22	1/15/22	2/1/22
BA.2.12.1	6/3/22*	N/a	N/a
Event Trough
Alpha	6/21/21	6/25/21	7/8/21
Delta	10/26/21	11/5/21	11/27/21
Omicron	4/4/22	4/6/22	6/3/22*
BA.2.12.1	N/a	N/a	N/a
---
---

So the fact that deaths are *still* declining more than 2 months after cases started to rise again speaks of a big change having happened.

Secondly, when we look comparatively at hospitalizations and deaths vs. confirmed cases, we see that with each successive wave, there are fewer hospitalizations vs. confirmed cases, and even fewer deaths.

The below chart norms the “Alpha” wave of winter 2020-21 to 1.00 for all metrics, and then compares successive peaks and troughs of subsequent waves with that respective peak and trough.

Note that, with one exception (hospitalizations due to Delta at peak), hospitalizations rose less and fell more than cases did during each successive wave after Alpha. Deaths have declined comparatively even more:

Event  Peak	Cases # (Thous.)	Hosps. # (Thous.)	Deaths # (Number)
Alpha	1.00 (251.8)	1.00 (117.7)	1.00 (3,393)
Delta	0.66 (166.2)	0.74 (87.4)	0.62 (2,117)
Omicron	3.21 (807.8)	1.31 (154.0)	0.77 (2,609)
BA.2.12.1	0.46* (115.7)	0.25* (29.0)	N/a
Event Trough
Alpha	1.00 (11.5)	1.00 (13.1)	1.00 (223)
Delta	6.17 (70.9)	2.71 (35.5)	3.68 (821)
Omicron	2.32 (26.7)	0.77 (10.3)	1.16* (258)
BA.2.12.1	N/a	N/a	N/a
---
---

At the peak of Omicron in January, more than 3x as many people had confirmed cases as compared with Alpha, but there were only 1.3x as many hospitalizations, and less than 0.8x as many deaths.

As of the most recent post-Omicron trough, cases were still nearly 3x as high as the post-Alpha trough, but hospitalizations made a new all-time low of .77x compared with the post-Alpha trough, and deaths - still declining as of a few days ago - are less than 1.2x as many as the post-Alpha trough.

Probably the recent relative decline is hospitalizations is due more than anything else to Paxlovid, which when taken promptly after the onset of symptoms, has been very effective.

When it comes to deaths, the causation may be one or more of many things: an inherent weakening of the successive variants, resistance due to the prevalence of vaccinations and/or previous infections in the population, more effective treatments in the hospital, or simply that the most vulnerable population already died of the disease, so there are many few extremely susceptible individuals left - or something else, or any combination of the above.

But, whatever the reason, COVID now has the approximate fatality rate as a typical  bad flu season. I expect new variants and new waves of COVID to continue, but appears more and more likely that it is ever so gradually transforming into an endemic condition.