The U.S. Economy Doesn’t Add Up

The Days Ahead:

• Retail sales, housing starts and earnings season

This Week:

• The U.S. economy doesn’t add up
• GDP and GDI are telling different stories again
• The Fed has a new index…we think it’s going to be important
• It has a better way to measure housing costs  
• The NFIB…still grumpy but still hiring and spending on capex
• A very good week for inflation

The U.S. Economy Doesn’t Add Up

When we look at the size and growth of the U.S. economy, we use the standard GDP numbers from the BEA. Past readers may recall, however, that there are several ways to measure the economy and the BEA uses two measures, Gross Domestic Product (GDP) and Gross Domestic Income (GDI).  

The “product” one just adds up the value of final goods. So, if a car rolls off the assembly line and sells to the dealer for $20,000, that’s the final product and that’s GDP. The sale to the consumer shows up in another measure called “final sales” but we can ignore that.

The “income” measure adds up all the income that the car produced along the way. That includes income for the tire manufacturers, the battery company and all the paid wages that went into making the car. It’s a lot more complicated than the “product” method, but the numbers should be the same.

Last year for Q1 GDP, we noted that a discrepancy of over $830 billion or 3.4% and that growth was either down by 1.4% (GDP) or up 2.1% (GDI) depending on which you read. We also noted that if there is a large discrepancy, it nearly always ends up with the GDP number being revised down. And so it was last year with Q1 GDP eventually marked down to -1.6%.

Well, it’s happening again in 2023. This year the latest Q1 GDP numbers are for 2.0% growth but the GDI number is at -1.8% for a gap between the two of 3.8%, which is greater than the 2022 gap. Those are annualized numbers for each quarter but if we look at a simple one-year growth of GDI from 2022 to 2023, GDI is down 0.8% and GDP is up 1.8%. Going back to 1947, there’s not been a year on year decline in GDI without a recession.

Here’s the chart in simple form.

The difference between the two was around $830 billion a year ago and is $400 billion now. The numbers will be revised. That $830 billion eventually whittled down to $277 billion but not until much later in the year. Here’s another chart showing the value of GDI over GDP.

When the blue line is above the black line, GDI is a bigger number than GDP and when it’s below the back line, GDI is lower than GDP.

Something clearly started to go astray after 2000. We won’t repeat all the reasons we dug into a year ago but it comes down to the BEA probably undercounting investment spending, which shows up in GDI eventually but can be defined incorrectly when the GDP numbers come out. Also, there continues to be discrepancies between different sources. The Fed records automobile growth at 11% in the last year while the BEA has it at 5%.  

It was enough of a mystery for the Cleveland Fed to write earlier this year a paper that asked:

Do Revisions Typically Reduce the Size of the Discrepancy between GDP and GDI?”

To which the answer, was, “Yes, with GDP moving in the direction of GDI”.

All this may seem flakey and way too much in the weeds. But it was discussed at the last Fed meeting as shown in the minutes published last week:

…several participants pointed out that recent GDP readings had been stronger than expected earlier in the year, while gross domestic income (GDI) readings had been weak. Of those who noted the discrepancy between GDP and GDI, most suggested that economic momentum may not be as strong as indicated by the GDP readings.”

So, we’ve been there before and our guess is that GDP growth will be revised down but not for several months. It won’t stop the Fed increasing rates in July, but it may mean that by September, the Fed will more clearly see the slowdown and pull back on further rate increases.

The Fed Has a New Index

Financial Condition Indexes (FCIs) are pretty common these days. All you need is to throw some data at a model, back test it and come up with a good title. Last we looked, several of the regional Fed banks have one and so do Bloomberg, Reuters and Vanguard. We probably missed another 50.

But there’s a new one from the main-Fed and it’s the first one they’ve ever produced. It’s called the FCI-G (for growth). Its feature is that it doesn’t try to measure whether financial conditions are easy or tight because that doesn’t tell you much about the economy. You can have slow sub-par growth in easy money eras (like the 2011-2020 period) and gangbuster growth in tight money periods (like 1996-1998).

What the FCI-G does is identify “headwinds and tailwinds to growth.” It looks like this:

Clearly the index says that the headwinds to growth accelerated very quickly in 2022 and peaked in December. They’re lower now but ticked up last month and remain historically high. The conclusion is that the headwinds to growth will be around for quite a bit longer.

The index also has some other interesting features.

One, it uses a lag effect, which we take to mean that any one measure in the index, say house prices, has a longer lag effect to growth. This makes sense. A 20% one-time drop in house prices may make a homeowner nervous but they won’t necessarily make a short-term decision about spending. After all, some spending is sort of committed. You have a vacation booked or a new washing machine on order. But the lag effect may be quite strong and purchases further out are either postponed or cancelled. Most FCIs drop their lag effect after one or two quarters but that seems way too quick to us.

Two, it uses the Zillow monthly index on house prices. Again, this makes sense because the Zillow index uses prices of houses actually sold or rented. The BEA method of calculating housing costs is, as we’ve discussed before, convoluted. It’s comprised of asking the question:

If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?”

It then tries to quality-adjust for the response, meaning if you answered $1,000, and live in a two-bedroom apartment next to an interstate, the BEA would then try and adjust for a large house next to the golf course. Finally, it divides the country up into six cohort samples but only asks the question for each cohort every six months. Much of the data is up to half a year old and it’s best to think of it as a long moving average of what rents are up to. Hats off to the BEA for trying to make sense of a complex problem and a large property market. But Zillow prices are a more immediate indicator about how confident consumers feel.

The Fed makes it clear that the new index is a “research product and not an official statistical release” and that it will be published on the 15^th of every month. But it also says it will be subject to the same blackout period that the Fed uses for the FOMC meetings.

Now why would the Fed go to the trouble of creating its first and very own FCI, subject it to publication delays in times of market sensitivity, and not use it for policy decisions? We think they will, but will be slow to acknowledge it.

Meanwhile, the index is sending the same signal which worried some people in the Fed when they discussed GDP and GDI. The economy is not as strong as it seems. The next publication will be July 27^th. We’ll keep an eye out for it.

It Takes a Lot to Please the NFIB

The National Federation of Independent Businesses (NFIB) is hard to please. Its members, small businesses, tend to dislike anything to do with taxes and regulations, are almost always in a better mood when there’s a Republican administration and get very cranky when the stock market weakens. Who can blame them? Fair enough. It’s hard to run a business, meet payroll, train people and keep prices competitive. But the NFIB matters a lot because small businesses employ nearly half the workforce.

Since its inception in 1974, the average monthly NFIB optimism index has been 98. It’s not been above 98 since December 2021 and last month it was at 91, up from 89 in April. That may seem low but there was some good news in the details.

First, credit conditions worsened in March after the failure of two major banks. Small businesses rely heavily on bank credit as they’re usually far too small to tap capital markets. But the latest reading returned to a much better number. Small businesses, at least, do not seem to face a credit crunch.

Second, 6% of companies said it was a “good time to expand.” That may seem low but it’s twice the level of a few months ago and about in line with the average for the last two years.

Finally, small companies are gradually increasing their capital expenditure. Over 52% reported making capital expenditures in the last six months. These are small companies so don’t expect big plants and factories. But a third bought new equipment, 21% bought vehicles and others expanded facilities or bought new buildings. Companies don’t usually do that if they don’t have some confidence in the future.

They plan to continue to do so as well. Here:

The chart shows that 25% plan to spend more in capital expenses in the next 6 months, up from 19% two months ago. This may seem low by past standards but markets look at what’s happening at the margin and these numbers are better than the recent equipment and non-residential investment reports.

The NFIB is notoriously cautious so the uptick in capital expenditure plans suggest they’re recovering well from the bank scares back in April. It also means that, while we may see a slowdown, a recession remains unlikely.

Getting Better

“A little better all the time, can’t get no worse” – The Beatles

The best news of the week was the inflation (CPI) report. The Fed has been on the “slay the inflation demon” for a while especially after the whole “it’s transitory/no it’s not” debate from two years ago. Headline inflation came in at 3.0%, its lowest level since March 2021 and down from a peak of 8.5% in March 2022. The core index, which excludes the energy and food prices, was up 4.8% which was its lowest since October 2021.

Most of the rise these days is due to shelter, which seems a down-market way to talk about housing, but that’s what they call it. It slowed on a monthly basis but remains about 8% above a year ago. Given how the measure lags (did we mention that?), don’t expect this part to let off soon. Other things like auto insurance, which we discussed a few weeks ago, are still running hot. But if U.S. consumers buy hard to repair and very heavy vehicles, auto insurance will remain high. Hard to imagine that a BMW i7 weighs about the same as a Humvee. Progress, eh?

Used cars are another troublesome component of inflation. But auction prices, which measures selling prices for fleet buyers like car rental companies, fell 10% last month and in the CPI fell 0.5% after months of increasing prices. The outlook is for more price decreases.

We’re big fans of long-term charts because our innate solipsism tends to think that current generations are seeing the worst, best, biggest, smallest of everything and so inflation in 2022 to 2023 is often described as a disaster of epic proportions. It’s not. Here’s inflation going back to 1950.

Yes, the spike is evident but look how quickly it’s come down. It took 15 month for inflation to rise from 1% (which is too low) to 9% (too high) and seven months to fall to 3%. In the long story of inflation, the last few years will be remembered, as:

Wait, you guys had a pandemic, closed the economy, 20% unemployment, introduced massive stimulus plans, raised import tariffs, saw a major conflict that kicked up energy prices and wrecked supply chains… and you bought inflation down quickly in a year? Respect.”

We think there’s more to come. Much of the disinflation in the last year had nothing to do with the Fed. Supply chains came back on line, commodity and energy prices fell and manufactured goods prices dropped. The delayed impact of the Fed’s rate increases will be around for a few months yet and we expect good news on the inflation side.

The Bottom Line

Good news on consumer and producer inflation, especially the trades services or margin components, won’t stop the Fed hiking on July 26^th. But it will probably be the last unless there’s some major risk on the jobs or inflation side. This week was the first time since March 2022 when both employment and inflation surprised to the downside. That provided a good back drop for with the S&P 500 up 2% for the week and up 17% year to date.

The Nasdaq-100 had a good week, up 3%. You may hear about a special rebalancing that is going on. It involves taking very overweight positions in the Nasdaq 100 index and lowering their overall weight. The top five companies (Microsoft, Apple, Nvidia, Amazon and Google) are 46% of the index and need to be around 40% for ETFs to meet diversification rules. We don’t think it will mean much, even though it’s the first time Nasdaq has had to do this mid-year. There may be some market noise for a few days.

The general good news on inflation meant the 10-year Treasury yield fell back to 3.7% having reached 4.1% just a few weeks ago. If earnings season goes well, and Delta airlines kicked off with some very good passenger, pricing and load factor numbers, we should see some good performance in coming weeks.

Subscribe below for our investment updates.

---

Art: Petra Cortright (b 1986)