The Days Ahead:

  • More housing; Jackson Hole symposium

This Week:

  • Yes, consumer debt is at an all-time high
  • But not as a percentage of household income
  • Credit card rates are over 20%, a 50-year high
  • Health care insurance is about to rise a lot
  • But it’s based on two-year-old data
  • Trade diversion or how to get round Russian sanctions
  • Why U.S. car exports are up 400% to 2000% in some countries

Programming Note: We’re off for a few weeks. We’ll get to think about the markets from a distance. There will be regular updates from Cerity Partners colleagues and the investment team here.

Consumers and Debt

One point about the current cycle is that the labor market, consumer activity and wages have all remained quite strong despite the Fed’s tightening. This was not always so. In the mid-2000s, consumers loaded up with cheap mortgages, cheaper home equity loans, credit card debt and auto loans. At its 2009 peak, the average household was spending 14% of its income on debt servicing alone and home mortgages stood at 100% of GDP.

The havoc from the GFC scared a generation of borrowers. People drew in loans, refinanced at lower rates and generally foreswore high debt levels. By the end of 2022, the average household was paying around 8% of its income on debt servicing and mortgages were down to 50% of GDP.

The lower debt and excess savings left over from the various stimulus programs goes some way to explaining the healthy level of personal spending in the last year. But that may change. Here’s the current level of credit card debt and the average interest rate charged as of the end of June:

Source: FactSet, 8/14/2023

Now, the average credit card rate is just an average. Plenty of people use credit cards like debit cards or carry a zero-interest balance and never pay any interest at all. Others may have introductory and preferential rates and all manner of incentives and loyalty cards. But the trend is clear. Credit card rates, which never really benefited from the zero rates of the 2009 to 2022 period, are at their highest level in a generation.

The level of debt at $1,262 billion is at a record high but as a percentage of GDP it remains low at 4.7%. In the pre-crisis period of 2008, it was 7% and it’s no higher today than it was in 1995. But sometimes it’s the rate of change that counts, not the level. If we assume that half of the credit card debt pays interest, then total interest payments will have risen from $92 billion to $138 billion in the last year. If we assume 75% of all credit card payers pay interest, then the amount goes from $151 billion to $208 billion. That’s equivalent to around 1% of GDP or 2.5% of retail sales.  

We haven’t seen a major uptick in delinquency rates. Total delinquency rates for credit cards, when the there are no payments made on the debt, were at 2.4% in March, way below the 7% we saw in 2009. But payments 90 days late stand at 6%, almost double the amount from a year ago.

What to make of it all? Well, credit card debt held by sub-prime borrowers fell and banks and delinquency rates of people who have credit scores over 660, are well below 2%, so it’s not as if banks are about to see a lot of write-offs. The New York Fed concluded their recent household finance report with:  

Despite the many headwinds American consumers have faced over the last year—higher interest rates, post-pandemic inflationary pressures, and the recent banking failures—there is little evidence of widespread financial distress for consumers.”

That seems about right but we’d point out that if there’s any hit to personal balance sheets, say, like a stock market correction or another leg down in housing, households will face record levels of credit card debt at interest rates higher than any level since 1972. That could limit the upside for consumer growth.

Health Care Insurance Time

We’ve written a few times about the strange way the Bureau of Labor Statistics (BLS) covers health insurance. It’s not, as you may expect, the cost of your insurance now compared to a month ago. There are far too many variations of health insurance to compare any two plans. It may be that one state requires certain procedures to be covered and another not, or the myriad ways deductibles, co-pays, out-of-pocket expenses, or prescription are covered. So comparisons are very difficult and some years ago the BLS gave up and used a system called a “retained earnings method.”

Every year, the National Association of Insurance Commissioners, which is all the 50 state insurance regulators, calculates how much the health insurance businesses received in premiums and how much they paid in claims. It’s not the same as profits, which include all sorts of expenses, like administration and marketing. It’s a simple model of “We took this money in to pay for health care and we’re taking it out, to pay for healthcare.” That’s fine except for one major problem: it’s out of date.  

Since October 2022, the BLS has been using the retained earnings calculated at the end of 2021. In 2022, health care premiums went up 8% but hospital and medical expenses went up 13.5%. That was mainly because people had all but stopped receiving anything other than urgent care in the midst of Covid-19 in 2020 but started to return for treatment in 2021. The result was that the earnings or “retained earnings”, for the health care companies fell from $151 billion in 2020 to $125 billion in 2021, a drop of around 20%.

The BLS then makes some seasonal adjustments and weights things like prescription, physician costs and hospital fees and comes up with an annual and monthly number. For 2023, we’ve been looking at 2021 costs over 2020 and they look like this:

Source: FactSet, 8/15/2023

The green line shows costs falling in 2023 between 10% and 30% on a year over year basis, and by 3.5% to 4.1% monthly, which is pretty much in line with the 20% decrease in retained earnings.

No, there’s no way your insurance or health care premiums fell. They’re more likely to have increased by around 8% which is what we see in the PCE Services Health Care inflation rate or the 3% that employers pay for health insurance. But it’s the system we have, so there we go.

This will all change in October because we have the new health insurance results for 2022 compared to 2021. For 2022, premiums increased 11% and hospital and medical expenses by 10%. The retained earnings went from $126 billion to $154 billion, for a rise of around 22%. And that’s the number that the BLS will plug into the inflation data from October 2023 all the way to September 2024.

That means two things.

First, the health insurance line in the monthly report will flip from around -4% a month and -20% on an annual a basis to +3% a month and +20% on an annual basis (depending on how they do the smoothing but close enough for our purposes). It’s roughly where we’ve used our Microsoft Paint skills to place an “X” on the chart above.

Second, the drop in health care inflation in 2023 has subtracted 0.04% from the monthly inflation report. That may not sound like much but when the Fed’s target is around 0.20% a month, every little bit helps. Starting in two months, health insurance will start to add around 0.03% a month. The Fed knows this, of course. But it’s one reason why the core CPI services ex-rents, which is what the Fed looks at, will start to jump.

Armenia Update

Back in February, we wrote about how tiny, landlocked Armenia with a population of 2.7 million people (and diaspora of 7.6 million) and GDP of $19 billion was crushing it in growth and exports. The economy grew 13% in 2022 and is expected to grow 6% in 2023. Exports grew 93% in 2022 and are expected to grow another 20% this year. The Armenian Dram has been on a tear, increasing by 30% against the U.S. dollar.

Much of its success is because exports to Russia increased from about $70 million a month to $270 million a month. Russia updated its trade statistics a few weeks ago and showed exports down from $154 billion a quarter in late 2021 to $101 billion now. At current rates, Armenia is one of Russia’s top trading partners although details of the bi-lateral data is either late or no longer available from the Bank of Russia.

Armenia is a rather complicated country and there are another three countries to this story. Georgia is to Armenia’s north and Azerbaijan to its east and west. There’s a road that links the two sides which runs through Armenia. Then there’s a disputed area of Nagorno-Karabakh, which was part of Armenia but lost to Azerbaijan in 2020. Since then, Russia has kept the peace between the two.

Meanwhile, Georgia is no friend of Russia given the war back in 2008 and has condemned the Russian invasion of Ukraine. Neither Armenia or Georgia joined the sanctions against Russia but because both are seen as special cases, neither are on the EU or US sanctions list.

In addition to Georgia, Armenia, and Azerbaijan, there’s also Kyrgyzstan. It too is not on any sanctions list or Russia’s “Unfriendly Countries” list, but has a very fraught relationship with Russia that is coming under more scrutiny. They’d probably like to keep out of sight and mind but the trade sanctions put them in a tough position.

In all four countries, trade has shown some massive changes over the last year and a half.

Now, we all know that sanctions are tricky to enforce. t’s a bit like whack-a-mole. One trade route is closed and another opens up. One of the EU sanctions is the prohibition of exporting vehicles valued over €50,000 or with engine sizes over 1900cc. Now there’s an elaborate US-Georgia-Armenia-Azerbaijan-Kyrgyzstan-Russia trade in new and used cars.

This the growth of U.S. exports of U.S. cars:

Source: United States Census, Cerity Partners

The growth is, well, amazing. Who would have thought so many people wanted America’s second-hand cars? It’s still a small dollar number but with very large growth and the four countries now represent 15% of all U.S. used car exports, up from less than 3% in 2021.

Much of the U.S. trade is in second hand or under-invoiced cars that don’t break the EU €50,000 limit. They are then repaired if required and shipped on to Russia at higher prices.

It’s not just happening in the U.S. Here’s German exports to Armenia, Georgia and Kyrgyzstan over the last few years.

Source: FactSet, 8/15/2023

You see the same exponential growth in exports from Poland, Sweden, the Czech Republic, the United Kingdon and Japan to these four countries. It’s basically trade diversion. Can’t get to Russia? Go around.  

No one’s particularly happy with it and all those numbers are in the public domain so it’s not as if anyone’s trying to hide what’s going on. There’s seems to be political preferences to allow smaller economies to grow on the back of trade diversion and see them cut back their historic ties to Russia, rather than being driven back into the Russian fold.

Does any of this matter to the wider markets and economies? Well, this week saw another devaluation of the Russian ruble, which is down 32% this year, even with an unchanged oil price. Russia is being forced to sell foreign assets that are not already under sanction and its budget and trade deficits remain under continued pressure. The sanctions may not be watertight but they’re enough to keep pressure on Russia and as long as that’s the case, they will continue to produce as much oil as it can.

In a roundabout way then, large trade jumps to four small Caucasus and mid-Asian counties, should keep energy prices low for the U.S. and EU.

The Bottom Line

We had a run of economic updates. Nothing much that would sway the Fed or us either way. Housing starts were down, building permits were flat, and industrial production was up, but a lot of that was from utility production and no real surprise given the heatwaves over the last six months. The Fed minutes confirmed the focus on “core services ex-housing” inflation, which is about 24% of the total inflation index and consists of transportation and food away from home. Those sectors all have a high dependency on labor and wages and it seems to bother the Fed that wages are running ahead of general inflation.

There’s also a widening gap between those who think there’s a risk of “real activity tilted to the downside” and those who “see significant upside risks to inflation, which could require further tightening of monetary policy.” Our assumption is that they will skip a hike in September.

Stocks have run sideways since mid-July. We’re tempted to put some of this down to not having a lot of new news on top of summer trading.

The 10-year Treasury made it to 4.3% briefly and it’s broken out from the 3.8% to 4.0% range for most of the time since March. We’d put that partly down to higher issuance but also to an easing of recession fears.

The Jackson Hole symposium starts next week and we’re unlikely to hear anything new from the Fed until then.

We’re off on a short break from this newsletter but please check our website for regular updates on all things markets and investments.

Subscribe below for our investment updates.

Art: Frances Bell (b. 1983)

Please read important disclosures here.