There Was A GDP Report?
We live in whirlwind times, and it all seems to coming to a head, all at once. So much is happening in such a short period of time that every week feels like lifetime. Do you even remember the stimulus talks, which played headline hockey with markets for weeks? That effectively ended only three weeks ago.
That this article is two weeks late should tell you something.
Believe it or not, an historical document came out at the end of October, the Q3 GDP report. Along with Q2, they represent the most rapid shifts ever in these important data.
In the media, this is being distorted by annualization of rates, which involves compounding quarterly rates over a year. Annualization is a way of saying, “if the economy keeps growing at this rate, then a full year would look like this.” But with rapid shifts happening in both directions, it’s useless and distortionary. Normally, it provides a quick way of comparing monthly, quarterly, and annual rates, but that becomes sort of meaningless in the current environment.
Real GDP is down 3.5% over 3 quarters, which for scale is -4.6% annualized. The previous growth trajectory was a tepid 2.1% per year, and we are 5.0% off that. Pulling out annualization in the Y-axis, that’s $734 billion off the trajectory. Real GDP levels are back to Q1 2018, before any effects of the tax bill, which were minimal anyway.
Looking at how that stacks up historically, it is the third worst 3-quarter period on record, Q2 2020 being the worst:
But of course, if you only look at the headline number, you are not seeing the full picture. There are many “K-shaped” recovery narratives, with parts of the economy doing well, and other parts not. This shows up more in the employment data along gender, race and income lines. You can probably guess who’s coming out on top there. But from the broad GDP level, the biggest splits we see are people substituting household goods and vehicles, especially used vehicles, for high density services that are either unavailable or unsafe.
Consumption, or PCE, was 68% of GDP in 2019, so it is the most important part. In the monthly data in the chart above, we can see this substitution happening. Of course there is also a ton happening in that red “Everything Else” line, and we’ll dig in there.
But the craziest economic chart of the pandemic still belongs to the personal income tables:
From wages, salaries, employee benefits, rents, dividends, interest, and income for proprietors, households have received $479 billion less cumulatively versus February 2020.
But they have received an extra $920 billion in government benefits for a surplus of $441 billion. On net, this was all saved.
Add to that $649 billion less spent on consumption, and another $150 billion less in interest payments, taxes and transfers, and households have saved a whopping $1.2 trillion extra since February.
These are all, of course, record moves. What happens to this savings bubble remains the key issue going forward. The personal savings rate remained elevated at 14.3% in September.
We see it all over: the data quality is extremely poor right now, and large revisions are becoming common. So, in the first place, remember that all this will undergo multiple rounds of revisions, and even in normal times, these can be large, especially in prices and deflated real numbers.
But these are not normal times, and agencies are struggling just to get the nominal numbers right, on top of trying to figure what is happening with fast-moving prices. So, we will look at nominal consumption and prices separately, and I encourage you to look at the prices especially as close approximations of reality. For historical data, we will look at real numbers. Like Guy LeBas, I believe it will take many months for this to sort itself out.
But more to the point, this is a site about publicly traded companies. Publicly traded companies deal in nominal dollars.
Consumption: The Incoming Situation
This is the most important part of GDP, not only because it is 68% of the top line, but also because it drives the second most important part, fixed investment. We are now 20 years into historically low growth rates in both these.
We’ll get to investment next. The incoming situation was far from “the best economy ever,” but rather a continuation of this long term trend that began around 2000. In fact, that long term growth rate has been declining steadily since 1980, with the brief pop of the late 1990s that dissipated quickly.
It is no accident that demand has slowed in the period where supply-side economics was ascendant. There are other, larger forces at work too, but we’ll get to that at the end.
But let’s zoom in on the last few years:
The Substitution Effects: Goods for Services
There are a group of services, mainly transportation, recreation, food and accommodations, that are not available or unsafe. A huge part of that cumulative $649 billion less households have spent versus February comes from here. But this is being offset by increases in goods purchases for the home, and vehicles.
This is the fastest shift ever in consumer preferences, even larger than the one during the last recession:
Services were 69% of consumption in 2019. In September, they were 66% of consumption. It took the entire cycle to go from 66% to 69%, and just a few months to go back. So let’s zero in on the high-density services.
These services were growing at 3.3% annualized in the incoming 7 months, right about the same as the top line PCE at 3.1%. So you see, these services are way down versus either February or the incoming trajectory.
Let’s look at the other side of it:
In contrast to the high density services, this group saw flat growth rates in the incoming months. Like everyone, April was terrible, but this is a true V-shaped recovery from that, a fairly rare occurrence.
But when we look at the dollar losses and gains over the last 7 months, they don’t come close to equaling out.
Let’s look at some of these substitutions starting with vehicles for ground and public transportation. For context, the largest transit agency in the country, New York’s MTA, has been releasing data on ridership and bridge/tunnel usage versus 2019.
The vehicles line was still on the rise in September, while the transit/ground transportation line has been pretty flat since the last two months. But even with the great vehicles performance since the April bottom, there were still large losses to make up from March and April, and they just got above zero in September.
Vehicle manufacturers, dealers and leasers built up a $21 billion revenue deficit in those first two months, and have been eating into it since. They finally got their heads above water in September. Meanwhile the losses in the transportation services keep piling up, another $10 billion versus February in September.
Now 2-3 reported quarters into this for most companies we’re beginning to see it show up on corporate income statements.
We don’t have the best look into the transit side, since most of this is public agencies and small taxi companies. Uber’s (UBER) top line was saved by surges in delivery, but if we split out the rideshares from the rest, you can see it is way down.
But let’s dig in here a little deeper, because it affects the Hertz (HTZ) bankruptcy proceedings. The incoming context:
So there is a totem pole. On top are used trucks/SUVs, then new ones. Both came off from their very high growth rates in 2019, but the trucks were still moving. In contrast, autos are flagging badly in popularity, used especially so since 2016. Anyone who has been to a Hertz lot knows that their fleet is largely made up of the bottom of the totem pole, used autos. Prices were deflating at almost 2% per year, and a flood of new inventory from Hertz looked to collapse those prices. This was the very grim environment for Hertz and their creditors, whose bonds were backed by that fleet.
But then this happened:
Used light trucks inflated at 13% YoY in September, and used autos 16%. Because inventories are still high, dealers are the ones making out, with their margin inflating at 19% YoY in September. So this gives a little hope that Hertz can exit bankruptcy, as the demand is all of a sudden there for all their used autos.
The next substitution is food and household supplies for food service.
After the March hoarding and April dip, food and household supplies have been pretty steady at 10% above February levels. The food services recovery has come off its July/August pace a bit, now down about 10% from February levels. Putting it into dollars, we see the gains in food and household supplies are eating into the losses in food services at a rate of about $2-3 billion a month, but that is a fairly slow pace.
Let’s dig into the food services a bit, because there’s a big split between fast food, full service, and school meals:
So fast food was able to creep above February levels by August, but the rest is doing poorly and looking very flat in September. This shows up even bigger in the October jobs report, and we’ll look at that next week.
Looking at some public companies:
Albertsons (ACI) saw the big early surge as people hoarded food and household supplies, which came off the next quarter, but still up 11% over the autumn. I chose Shake Shack (SHAK) just because they are domestic. You see they are off their 2019 autumn quarter quite a bit, but unlike full service Darden (DRI), coming back fast.
Next up, people are stuck in their homes much more than they have ever been. They are working from home, and also not taking vacations. So people are spending more on furniture, appliances, home/garden, as well as recreational goods substituting for all the recreational and travel services they are not using.
That services aggregate remains down 43%, while the goods aggregate bumped even more in September to up 15% from February levels. But the losses piling up in that green line are giant, almost half the cumulative reduction in consumer spending since February.
Turning to representative public companies in the home and recreational goods aggregate, we see they are all up sharply in the summer quarter.
But the flip side of that is grim.
The question moving forward is how durable these shifts will be. Investment in manufacturing tells us manufacturers believe it is temporary:
Consumption: The Rest of It
Back to our original chart:
So this section is about that red line, in aggregate just above February levels. But as I promised, there’s a lot going on under the hood there. That red line is 59% of consumption, mostly because it includes the two largest categories, housing and health care, which together are 33% of consumption. All together, the cumulative deficit in this aggregate is $121 billion versus February levels.
Starting with housing, it is acting like nothing is happening:
Moving to health care, this is also a giant category at 17% of PCE. But if we add in pharmaceuticals and other medical goods, homes for the elderly and rehab (which get listed under social services), and net non-profit health care services, that is 24% of PCE. Almost a quarter of what American households spend is on health care in some manner. It has been a long road back, but it is almost there:
But this is a giant category, so the area above that line represents a cumulative loss of $159 billion dollars for the sector. These losses have been concentrated in doctors’ and dentists’ offices and especially hospitals:
HCA (HCA) has the largest hospital network in the country. They finished their summer quarter just below their autumn quarter. But the winter and spring quarters were just terrible.
But those services has been offset by some of those goods categories a bit, like pharma, especially non prescription, up 9% from February levels after dipping hard in April.
Finally, the two large goods categories that are doing very poorly: clothing and gas.
Clothing had a little back-to-school boost, but is still categorized by falling prices through the entire chain, all the way down to fiber.
I see a little bit of confusion about what is meant by investment here in the GDP report. It does not refer to the type of investment we talk about in finance, which for individuals is categorized as savings in the economic sense. Investment in the economic sense is goods and services purchased by companies for buildings, equipment or IP.
Like consumption, long term real investment growth is at its lowest level ever in the tables, which only go back to 1967.
As you see, the 2017 corporate tax cut did nothing to change this. Consumption growth was low, so there is not a lot to invest in. Again, we see the peak around 2000. The surge in buybacks was because there was not a lot of CapEx that was attractive, so the tax savings were returned to shareholders where, it mostly stayed in personal savings, and did not stimulate either investment or consumption.
Zooming in on the last few years, contrary to the promises of the tax bill, investment growth was a little below the incoming trend.
Outside of IP investment, the overall picture was fairly weak coming into the pandemic:
Fixed investment was growing at an incoming rate of 2%, which is pretty tepid, and is now 3% off that pace. Overall, that is a nice recovery in Q3. It is concentrated in trucks, IT equipment, warehouse construction, but especially residential construction.
This is a pretty wild ride and I don’t think it’s over. Residential investment was already hot coming into the pandemic, and is now up 8.4% over 3 quarters, with a huge recovery from the Q2 dive. But what we find in the splits is not what I expected considering how the media is covering the housing market.
The big winners here are apartment buildings and improvements, but new single family home construction is still below Q4 2019 levels, despite prices inflating at 5% YoY. Moreover, none of this is showing up as CapEx on the recent reports of homebuilders or apartment REITs.
So honestly, I’m not sure what’s happening here, because the data has a lot of contradiction in it.
The picture in nonresidential is less pretty, mostly owing to structures.
So that’s a big V-shaped recovery there. Especially strong were IT equipment and trucks. Boeing (BA) even sold a few aircraft.
The strangest thing in the investment table is IP investment, which has been the strongest thing in the tables for a decade
Normally, inventories have little effect on the top line of GDP. But when we are going through cycles of stocking and destocking, it can have an enormous effect. For example, in Q3 2018, companies raced to fill up warehouses in anticipation of tariffs in the fall, some of which never arrived. But the result was that 36% of the nominal GDP growth in the quarter was from inventory buildup.
So we have just gone through a giant cycle of destocking that began with the supply chain disruption in China in January which continued through Q2. Then in Q3, we saw a huge amount of restocking. In all three quarters, inventories are having an outsized effect on the top line
To make sure that the gross product remains domestic, imports are subtracted from exports and the difference, the trade deficit, is added or subtracted from GDP. The trade deficit surged in Q3.
Both imports and exports came into this on a downward trajectory from the trade war, imports more so. Imports got hit harder at first because the pandemic closed down China earlier than the US. But with the surge in goods demand we saw in the consumption tables, imports had a bigger recovery in Q3, so that’s what made the deficit spike like that.
But trade is still doing terribly, with imports down 9% and exports down 18% over 3 quarters.
GDP measures what is being bought and who is buying it. It does not measure income, part of the national income accounts, which is who is getting the money, and from whom. So the GDP report only measures a part of what governments do — the goods and services they purchase. Employee salaries and benefits only get counted when the recipient spends or invests the money. If they save it, it never goes into GDP.
So the GDP report actually tells us very little about the role of government right now, because of that giant income surplus households have gotten from government benefits is all on net still sitting in savings as of September.
But the national income tables do give us a much fuller picture:
- Transfer payments, mostly benefits are still elevated at the Federal level, but way off Q2 levels.
- But transfer payments at the state and local level were still rising in Q3.
- Fed consumption remains elevated in Q3, but off Q2 levels. This is from all the PPE, ventilators, Project Warp Speed, etc.
- But at the state level, consumption is starting to come off as states look for places to trim.
- The jump in the Federal Other category was aid to states in Q2, and business subsidies in Q3.
- But again, the state and local Other is coming off as states look to trim.
As you may be able to tell, I’m a little worried about state and local spending, which has mostly held up so far. But they cannot run deficits. We do not get tax receipts data for Q3 until next month, but this is what was happening through Q2, excluding Federal direct aid:
Why Was Everything So Slow This Cycle?
Let’s look at that long-term GDP growth rate chart. I’ve added in the per capita line, so you can see the effect of slowing population growth, which is the first part of the explanation.
So about 14% of the reason growth has been slow in the last two cycles is very simple, and that is slowing population growth. The per capita line is pretty steady in the 2-3% range from the late 1960s until 2002, when both line begin to fall off.
But it is also an aging population, and once people hit 55, they are much less likely to be in the workforce. The oldest Boomers turned 55 in 2001, and the labor force participation rate has been declining since.
The blue line is the percentage of US population over 55, just above 20% after the adjustments from the 2000 census. That group is now almost 30% of Americans. The labor force participation rate peaked in 2000. So the second part of our explanation is that we were back to late 1970s levels of labor force participation, before all the Boomers had entered the job market, and before the shift of women in the workplace had finished.
So a larger portion of the population is living off an income that is lower than their working-days income, so they are consuming less. They maybe even have a negative savings rate, living off of savings. So this group stimulates the economy to the extent that they can by consuming all their income or even more. But their income is relatively small, so the overall effect is a drag on consumption.
The final bit of demographics involves increased life expectancies. Workers work the same number of years as they used to; the average retirement age remains around 65. But now, people are alive much longer, and workers have to self-insure against running out of money in their 80s or even 90s. There’s a good chance you are here at Seeking Alpha for this very reason.
That line is the average life expectancy at birth minus 65 — a rough estimate of retirement years that can be expected at birth. It is now over double what it was in the late 1960s, so workers must save over twice as much for retirement. In effect, they are deferring consumption for many decades.
So a big chunk of the problem gets explained by demographics. Japan and many European countries have the same issues, though more advanced. China has the largest demographic issues from 70 years of Communist Party social engineering, but they won’t show up for a couple of decades.
The next part of the explanation of why we are over-saving is income inequality. The ultrarich save almost everything they get. There are just so many private jets and islands to buy. The poor spend every dime they get on essentials and save almost nothing. Most people are between those two extremes, but the basic relationship holds; the higher your income, the more likely you are to save the marginal dollar of income.
A Gini coefficient is a way of measuring the distribution of a set of numbers. Its main use in economics is to measure income distribution — the GINI Index from the World Bank. A GINI Index of 0 means everyone gets the same income, and 100 means 1 person gets everything. These are obviously theoretical extremes, and sane values are between 20 and 55. As you can see, the US is the highest among advanced economies, and more on par with Russia, China, and developing countries. The US Gini Index hit the low 40s in the mid-1990s and has been there since:
The next part of the explanation is supply side policies, still in effect long after the problems they solved were with us. For this we have to back up to the late 1960s, as women started entering the workforce in larger numbers, and the oldest Boomers were in their 20s. Let’s look at that long term consumption chart again.
In the 1960s and 1970s long term real PCE growth was pretty steady around 3.9% annualized, keeping long term real GDP growth above 4%. All this consumption caused a surge in capital demand for investment. But there was not enough savings to fund the investment, so inflation spiked as demand outstripped supply. Interest rates had to rise to entice more savings, but that also puts a damper on investment.
Add in 2 oil shocks, and you get a perfect storm of inflation and high rates coming into the 1980s.
So the basic problem was that the marginal propensity to consume was too high because we had such a large, thriving middle class. Add in the oil shocks for even more nominal consumption. Companies were starved for capital to try and meet this demand, and only high interest rates could persuade people to save.
So the policy response was twofold. The Fed raised overnight rates to unheard of levels, getting as high as 22% in 1981. This crushed the economy and led to a double-dip recession. But it also crushed inflation and brought it back to more manageable levels by the time the second dip ended.
At the same time, the Federal government pursued loose fiscal policies — tax cuts paid for by debt — to counteract the draconian monetary policy happening. These policies were designed to raise the supply of capital at lower interest rates.
- Reduced nominal rates on high earners who will largely save that bounty.
- Special rates for capital gains.
- Tax advantaged savings accounts for retirement, health care, education, etc.
This worked great. The economy grew at a nice clip, and inflation was tamed by the end of the 1980s.
That’s real GDP growth broken up into its two components, nominal GDP in blue and inflation in red. Inflation remained a little high, but not nearly as bad as the beginning of the decade, and rapid nominal growth made up for it.
But then the shifts in demographics we discussed began changing the underlying context. Instead of having a population bubble of new workers driving demand too high, we began to have forces encouraging too much savings and oversupply of capital.
That’s how this happens:
Interest rates are a price for money, and like all prices, they are determined by supply and demand. The high interest rates in 1981 told us that capital demand growth was outstripping supply growth, and we needed more supply. The exceptionally low rates of today, even pre-pandemic, are telling us that supply growth is outstripping demand.
We need more demand.
While we backed off supply side slightly during the Bush I, Clinton and Obama administrations, the Bush II and Trump administrations reversed that. So these policies designed to solve the problems of 1980 are still with us in 2020 when we have the exact opposite problem.
Why Did I Just Go Through All That?
I am measuring the savings bubble differently than Ian Shepherdson, Chief Economist at Pantheon Macroeconomics, whose tweet is screenshotted above. I am comparing savings to the February monthly rate of savings, while he is comparing to the overall February level. We agree however that it is giant regardless of how it is measured.
In addition to that cash bubble there is:
- An additional $1.2 trillion in bank reserves since March.
- An additional $1.3 trillion in Treasury cash being held at the Fed since March.
- An additional $1.1 trillion in nonfinancial business cash and equivalents in Q1 and Q2 2020.
Add to that $5 trillion in T-bills earning a negative real return, and you have almost $10 trillion sitting around doing nothing. The most important of them is the household bubble, as others will react to what households do.
Ian Shepherdson thinks that households will spend that bubble once we are all vaccined up. That is the negative savings rate I mentioned in the intro. I am less convinced. It all depends on who is holding that bubble. If it is largely in the hands of high income households, it will likely remain in savings until they retire and start spending it.
If, however, it is evenly distributed or concentrated in low income households, it will stimulate the economy as Shepherdson suggests, and may even spike inflation. That is not even close to a problem right now, because the Fed can very easily raise rates or roll off balance sheet if it persists.
I suspect that many low income households are out of savings, however, and more will be by the time enough people get the vaccine. My guess is that the primary distributional effect of the pandemic is that the rich got richer.
We don’t have hard data on that yet, but there have been many indications in real time data that the reduction in consumption is coming largely from high income households:
So the reason I went through the whole song and dance of the previous section is this: we already had too high a propensity to save the marginal dollar of disposable income coming into the pandemic. One of the reasons is income inequality and the supply side policies that have encouraged it. If my guess is right, the pandemic has just deepened that inequality in a big way, and set low income households back years.
But if Shephardson is right, there will be a Biden Boom.
What to Expect from a Biden Administration
The President’s tweets aside, Joseph R. Biden has won the 2020 Presidential election and will be inaugurated on January 20. There will be changes.
One thing that seems to have escaped media coverage of what sort of administration we may see is one obvious fact. Democratic primary voters had a choice between progressives and even a democratic socialist, but chose the most conservative candidate in the field. His primary campaign insisted repeatedly he could find common ground with the GOP, and he still won.
The Democratic coalition remains a center-left coalition, and that has not changed. “The Squad” is popular on Fox News and Twitter, but Alexandria Ocasio-Cortez will not be running the House; Nancy Pelosi will be.
And moreover, there’s Mitch McConnell in the Senate, and he remains the smartest and most ruthless man in Washington. Even if the Democrats win both Georgia Senate runoffs, which looks unlikely, they will still not have enough votes to kill the filibuster, with Joe Manchin of West Virginia already dissenting on the idea. So 60 votes, even for bathroom breaks, will remain the norm in the Senate, as it has been when there’s a Democratic President since 1994.
Filibuster use has increased dramatically, but especially when Mitch McConnell has been the GOP Leader in the Senate. That chart really understates the issue, because many things are never brought up because they are majorities, but not close to the 60-vote supermajority.
At best we are looking at lots of compromises for any legislative action Biden may want to take. The Senate may just say no to everything, as they did to Biden’s former boss.
So legislatively, I do not expect much. There will be a stimulus package, but it will be much closer to the Senate’s $500 billion than the House’s $2 trillion. After that, Biden will try very hard to find places for compromise like infrastructure, but still may fail.
But that still leaves Biden with wide latitude to act without Congress, as we have seen in the past 4 years. Here’s a few important things to expect:
- Repairing the relationships with traditional allies, and the strengthening of international organizations like NATO, WHO and WTO. We will rejoin the Paris Accords, and, if dreams come true, TPP.
- A big part of this will be a quickly negotiated end to the trade war with our allies. Hopefully that involves some give on aircraft disputes, which preceded Trump.
- China is a different story. Biden will try and use these repaired alliances to counter China. Passing TPP would help with that. Everyone else wins if China is balanced in Asia, but the big winners here would be the south and southeast Asian countries if he is successful. It’s unclear how Biden will handle the managed trade in the Phase 1 agreement, for which there was never a Phase 2. He is a multilateralist, unlike Trump..
- Rollback of Trump deregulation, especially in environmental regulations. This will take time.
- Rollback of many Trump executive orders, and a large reduction in their usage. These last two will keep lawyers gainfully employed around the clock.
- Restaffing of the departments, filling many roles left empty by Trump with expertise.
- A national strategy on COVID that balances the limitations of Federalism with the need to control a raging pandemic.
Too often the pandemic is discussed in political or economic terms. But this is at root a public health problem, and since we have failed to solve that, the economic and political problems will persist. A quick look at where we are.
The national case rate has gone parabolic.
- 19 days in a row with new daily records in the moving average case rate.
- 130,000 per day which is 380 per million. That means 910,003 Americans got COVID last week.
- The positivity rate is up to 9.2%. Testing is not keeping up.
- The case growth rate is at 26% per week
This is the most widespread the problem has been, with every region beginning to soar.
I see no appetite for another round of serious lockdowns, so this will just continue to get worse through the winter. Thanksgiving will likely be a locus of renewed infection. Deaths, which tend to start rising a month after cases, are back over 1000 per day in the moving average.
And the problem is not isolated on this side of the Atlantic.
What I am trying to say is that the near term outlook is grim. We will have the first doses of the vaccine in December through Emergency Use Authorization, but they will be in relatively small numbers at first. Pfizer (PFE) says they will have 50 million doses by the end of the year. But vaccination requires 2 doses, so that is 25 million people that can be vaccinated worldwide. The US will get a big chunk of that but not all. There will be rationing, which will become a political football. The anti-vaxxers, and the folks who believe it is all a plot perpetrated by Bill Gates to implant a microchip in everyone will have their say. This will not go smoothly.
Pfizer says they will have 1.2 billion doses in 2021.
The good news is that since Pfizer was so successful in replicating the COVID spike protein, the likelihood that other vaccines were also able to do so is greater, so we will more have more than just the Pfizer numbers to add to that.
There is also the huge logistical problem of these new mRNA vaccines, which must be shipped and stored at ultra low temperatures. Many areas will need Federal funding to help with distribution.
This is going to be a very tough fall and winter.
But beyond that, once enough people have gotten vaccinated worldwide, and people are willing to go into crowded spaces again, then and only then will the recession begin to end. At that point what happens to that savings bubble will be determinative.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.