11 min read

Business Cycle Watch #3 Indicators' divergence

What was seemingly inevitable became real, the Treasury yield curve has inverted. Even though, almost all coincident US indicator signals are in green!

The economy is still in a reflationary mode, but we're most likely coming to the end of the expansionary phase of the cycle. As 2s10s continues to plummet and the Fed keeps affirming it's extremely hawkish stance, the chances of a coming US contraction grow bigger and bigger.

Mixed signals

March has passed, so we can look at final numbers of the mother of all indicators - GDP. After mediocre Q3, the last quarter of 2021 continues a strong recovery from the shortest-ever recession. Over the period, the US economy expanded 6.9% on a saar basis (numbers on the graph below are not annualized), and last month revisions pulled the number -0.1% from its original estimate. Looking at the whole 2021, growth of the American economy was the highest in decades. The last year that comes closest to the '21 growth of 5.34% is 2003 with a 4.3% growth rate.

If we look at the other countries, Q4 GDP suggest that the recovery phrase is ending, and, at best, we are entering the old times of 2010s status quo. Chinese GDP comes to the level of late 2017 and the Q4 growth of the Euro zone is similar to that one of 2013 when Europe was starting to slowly heal from the euro crisis.

Q4 2021: US 1.68%, Euro area 0.26%, Germany -0.35%, France 0.71%, Japan 1.13%, China 1.6%.

On the trading floor, the bear market does not continue. In the last month, SPX grew ~5.8%, and Nasdaq composite even 9%. Clear communication from the Fed pushed markets, including equities, to discount before interest rate rises even started. Hence, low sensitivity to the last rate hike, and even a somewhat positive reaction since many were predicting a 50 bps increase.

Year-to-date change of SP 500, though, is still negative (-3.8%), only energy and utilities SPDRs have risen this year. A rapid one-month growth in XLY (consumer discretionary) suggest either a retail-fueled economic expansion or simply "buying in a dip" Tesla and Amazon stocks.

A slightly different picture comes out when we look at XLF (financials) and XLU (utilities) funds. Financials trade is typically a bet on reflation, as indices tied to that sector closely follow the credit cycle. By contrast, utilities, by providing a low but stable adjusted dividend and having a very low beta, reduce the overall risk of a portfolio. As the price of XLU rapidly increases, the higher uncertainty in the markets, and the higher chances of a bear market. If we compare price of both ETFs, we can get a feeling of a reflationary sentiment in the markets. The two graphs below show that the equities, collectively, can't decide if we're in a boom or a downturn.

Even though the hiking cycle has started, financial conditions haven't deteriorated at all. 3-month 10-year Treasury spread as well as the FRA-OIS [A] spread are both on a very sound level, yet, bank stocks aren't quite exploding. That hesitation may be explained by what is happening with other Treasury spreads. 2s10s have just inverted 1 basis point last Monday.

While my constant talk about inversion may be questioned as the mechanics of how such yield reversal contributes to a recession is unclear, the empirical evidence of 2s10s predicting capabilities is astounding. Out of seven last inversions, six were followed by an official NBER recession, and the only outlier was 26th May 1998, the time just after the Asian financial crisis and just a month before the LTCM problem started showing up[1].

Blue lines = 2s10s inverts for the first time. On April 4th, the 2s10s closed at -1bps.
2s10s inverts for the first time Recession starts Days (roughly)
18 Aug 1978 Feb 1980 530
12 Sep 1980 Aug 1981 320
13 Dec 1988 Aug 1990 600
26 May 1998 - (almost) - -
2 Feb 2000 Apr 2001 420
27 Dec 2005 Jan 2008 730
27 Aug 2019 Mar 2020 180
1 Apr 2021 ? -

Despite a recent surge in the 10-year T-note yield, which saves interest margin profits, the near future of the lending business is definitely not optimistic. The first sign is, and probably the most important one is, a very hawkish forward guidance from the Fed. Second, a very mediocre performance of XLF. Third, yield curve inversions: eurodollar at roughly 1 year-to-maturity, Treasury at 2-year YTM and IRS[B] at 3-year YTM.

It can be argued, however, that a recent rise in the 10-year yield reflects a confidence about an upcoming growth and the effectiveness of the Fed. The argument gets a bit weaker when we look at global long-term yields around the world. An upsurge in 10-year USTs was accompanied by rising yields in other major areas as well, including the Euro zone countries, Japan and China. Please, note that BoJ and PBoC haven't been much hawkish lately while the ECB, despite all desire to hike, have no choice but to wait out the situation in the east. A synchronized growth? It may be the case for this quarter or even the whole year, but what then?

Eurodollar curve saw that coming in Dec '21!

That is essentially the whole thesis that I have been about writing since early February, which is a long-term deflationary judgement (in monetary terms) of global markets despite high CPIs. Supply shocks and fiscal action surely affect prices, but they're always temporary and don't break the paradigm of private sector growth.

This trend, although not always in the same form, play outs across many developed markets. UK's gilt curve is completely flat[C], zombie-full[2] Western Europe is in an energy and geopolitical crisis, and Asian economies have a flavor of the same recurring problem, too. (More on Japan below.)

However, when we look at other non-market economic data, then we are led to believe that things are going great, which... may actually be the case. Non-farm employment increases, high JOLTS numbers, Markit PMIs almost 60, still no significant contraction in retail. Everything seems great.

The question then arises, which indicators do we value more? Market-based, forward-looking information or surveys, which we aren't sure the quality of. Data that comes from the collective of investors with the skin in the game, or the collective of CEOs that don't even get paid for filling in those papers.

In case of an indicator conflict, I would advise choosing the markets.

Don't get me wrong, both kinds of data are extremely useful. Many diffusion indices that gauge business' sentiment can quite accurately depict the current phase of the cycle. None of them, though, can predict a sudden stop in liquidity or other high-skew event. Markets can. Or at least they can discount, and show that prospects of such happening are in fact probable.

After all, surveys will always stay surveys, and then, most of them represent the past or the present. Not the future [3]. Though, markets always do look forward.

There is also one more bit of the Treasuries market that catches my attention, which is the short-end of the curve. Throughout 2021 there were many times when short-term Treasury yields went below the Fed's reverse repo facility rate that sets the return of collateral that can be obtained from the Fed's balance sheet. These drops were both counterintuitive and worrying. Counterintuitive because in a perfect market setting, where every player is equal, Treasury buyers bidding should never exceed the price equal to the RRP rate.

Why buy bills for more than the rate the Fed offers through its standing facility? There can be three explanations for that.

  1. A high demand from non-eligible RRP buyers;
  2. A high demand for on-the-run bills that the Fed may not actually have many;
  3. A general bills collateral shortage.

The possibility that it is the third option is worrisome. A T-bill shortage means a collateral problem, which in effect is a monetary problem.

It seems that, in 2022, those bills are still in high demand. Increasing the RRP rate bar only showed the magnitude of this issue. A difference of over -10 bps in yield between a 1-month bill and the RRP shows that the imbalance is much bigger than initially thought.

Although, it is not the end of the world, and fears of an imminent recession shouldn't be exaggerated, warnings of the yield curve alarmists are definitely not unfounded. If the gloomy prophecy of negative 2s10s fulfills once again, we still have about a year of time for the worst to come (see the table above).

Europe on the verge of a commodity crisis

The war in Ukraine is setting up the Europe, Africa, and perhaps even whole global supply chains into a state of emergency.

The EU relies immensely on the russian gas and oil. As many as 30-40% of European energy imports come exactly from that location. At the same time, western European leaders are feeling a bottom-up pressure coming from the forces of democratic order to pivot their international policy towards the aggressor. The EU is in the dilemma of whether to impose energy imports on the russian state and push it into an autarky and poverty, by also inflicting enormous damage on their own economies, or to pretend that "oligarch sanctions" will be enough.

The question is much more geostrategic than economic, as the security structure of the whole Europe is being questioned. Yet, businesses and markets do feel anxious. A possibility of future energy rationing of German produces was reflected in two major surveys – ifo and ZEW.

Ifo Business Climate Index measures the seize-weighted and GDP-weighted opinions of businesses on the current business situation and expectations for the next six months. Last month, the business expectation index dropped a whopping 13.3 points. If we exclude periods of GFC and the covid crisis, then that's the largest decline in the whole time-series that goes back to January 2005.

ZEW sentiment index that polls economists and bankers, fell to -39.3, which means that, if we exclude indifferent votes, almost 40% of correspondents see economic conditions deteriorating in the near future. In one month, the indicator went from 54.3 to -39.3. A -93.6 drop, which is the fastest fall ever recorded. Second fastest is only the March 2020 reading, with a change of  -58.2.

Those surveys may well overestimate a potential upcoming downturn, as the European stock market isn't much concerned about instability in the international-strategy domain. Both the broad European index STOXX Europe 600 and Germany's DAX gained in March about 4%.

Whether stable equities is a rational evaluation or a benign neglect of the commodity crisis is difficult to say. There is a hope that things aren't as bad as we may think. The EU carbon permits market haven't exploded (-0.5% YTD) in the 2022, yet. Neither have Newcastle coal futures (+2% YTD). Despite that, the price of black gold have started to increase in the first days of April as the EU abandons russian coal imports and seeks to substitute it with the one that comes from Indonesian mines. If the same fate befalls other energy sources as well, then the next winter may be the biggest European energy crunch in many years.

Wheat and oil futures skyrocket, while carbon emissions level off, at €80.

Asian problems

While US faces finance worries and EU faces energy worries, Asian economies experience yet another set of problems.

In the last of its monetary policy meetings, the Bank of Japan have decided to keep the massive stimulus going. This resolution slightly surprised markets, as it had been thought that the BoJ will jump on the hiking bandwagon with the rest of the major central banks.

A low inflation rate that haven't even reached 1% yet (y/y CPI) gives some space for Kuroda & Co. to continue their current policy.

Japan's economy is still in the midst of recovering from the pandemic's impact. What's important for us now is to support the recovery by maintaining easy monetary policy.
– Haruhiko Kuroda

As the governor of the BoJ recently said, the Japanese economy still hasn't recovered from the pandemic. Industrial production, is at a much lower level (95.8) than the base one of 100 that was set in 2015.

The overall sentiment of households and corporates also isn't very bright. The Watchers survey is a set of diffusion indices that is created by the Cabinet Office of the Japanese PM and measures activity of businesses and regular people. Think, a Japanese PMI issued by the government, that also gauges retail. The main index that combines all surveys into a single number is currently at 37.7. By contrast, the average of 2021, the first year of the recovery, was 45.

Watchers' in February: Total = 37.7; Housing trends (not in the plot) = 33.7; Housing retail = 35.6; Corporate trends = 43.1; Employment (not in the plot) = 52.1. Employment was the only index that increased m/m.

In addition to the weak manufacturing and grim sentiment, lending in the last 10 months have been dramatically low. The current y/y growth rate of loans given by banks is at 0.36%, while the 12-year mean is 1.8%. Retail sales are also depressed. The latest year-over-year change of retail is -0.8%, over 2 percentage points below the long-term average.

February Sales: Total = 6.1%; Wholesale = 8.8%; Retail = -0.8%.

The Japanese problem is a problem of growth, which isn't a new phenomenon, as the country have struggled to generate an economic growth in decades. BoJ is trying to stimulate with its extension of QE and the yield curve cap at 25 bps. However, the post Bretton Woods experience suggests that any kind of fixed caps, floors or other pegs almost always face the market's backlash. The price for the 10-year interest rate cap are capital outflows that devalue yen in the foreign exchange market.

An external devaluation may hurt as energy commodities increase in price and Japan imports most of its energy from foreign countries. Albeit, preventing the cost of the government debt to rise in a country with a ~225% debt/GDP, and engaging in another BoJ intervention to support the yen is probably a more sensible resolution.

Japan is not the only Asian economy that faces some headwind. There is also China, but I will save an analysis of that country for the next one in May.


[1]: "In June 1998, LTCM racked up a 10 percent loss, their largest one-month loss to date". Source.

[2]: I am referring to "zombie companies" with debt service coverage ratio less than 1.

[3]: This is also true for employment numbers and many other non-survey but also non-market data.

Some more charts.

[A]: FRA-OIS spread.

Source: Bloomberg.

[B]: USD IRS curve.

Source: Bloomberg.

[C]: Gilt yield curve.

Source: WSJ.