LEI Sep-2023
20-Oct-2023. US Leading Economic Index. We’ve just got the reading for September and the Leading Economic Index (LEI) fell 0.7%, after falling 0.5% in August. This decline represents an annual change of -7.84%. These are of course (historically) recessionary levels. However the LEI annual change has not deteriorated for 7 months... which can be interpreted in favor of stronger economy narrative and an adequate H4L (higher for longer) environment. Another argument in favor of H4L is the long Fed pause expected by the market. Similarly, ISM Manufacturing, which has been increasing since March, also supports H4L (for now). See chart 1 and the "H4L" writing. It was similar in 2006-2008. Both the FED’s pause (15 months), the LEI not falling more (about 10-12 months), and the slightly rising and not falling ISM Manufacturing (about 12 months) - then similarly indicated H4L. Chart 2. It took then a long time ... and did not end well.
LEI vs S&P500
24-Nov-2023. Leading Economic Index vs S&P500. Part 1. This week we got the reading of the US Conference Board Leading Economic Index (LEI) for October 2023. The annual change in this index in October 2023 was -7.64%. The year-on-year change of the S&P500 index in October 2023 is +8.3%. The chart shows the comparison of the S&P500 with the LEI. What may be worrying now is the lack of LEI rebound from current levels. Typically the stock market is faster and current S&P500 levels suggest a stronger economy and higher LEI levels. This relationship between S&P500 and LEI is especially useful after big LEI moves, and the LEI drop from April 2021 (+11.67% YoY) to April 2023 (-8.13% YoY) is a big deal. So, which index is wrong? Can’t both be right.
25-Nov-2023. Leading Economic Index vs. S&P500. Part 2. Calculating the annual change in the S&P500 based only on the index level on the last day of a given month could be a bit arbitrary. Therefore, I also calculated the yearly S&P500 change based on the average level of the index in a given month. Comparison in Figure 1. Generally, there are no major differences and the conclusions will be the same, but to be precise, the maximum positive difference was 15 percentage points (Feb-2010) and the maximum negative difference was 12.9 pp (Feb-2020, as a reminder S&P500 from 19-Feb-2020 (top of the cycle) to 28-Feb-2020 dropped 12.8% - this obviously affected the calculations for that month). Moving to monthly averages, we obtained a more smooth S&P500 YoY. The difference to the annual LEI change in October 2023 is even greater... - see chart 2. Now the S&P500 YoY (for average monthly levels) is +14.6% (in the version based on prices from the last day of the month it was +8.3% (the difference is due to the decline in the S&P500 at the end of Oct-2023 (vs. the monthly average) and an increase of S&P500 at the end of Oct-2022 vs. monthly average). Of course, the conclusions are the same... that the S&P500 either: (i) front-run the upcoming recovery in the US economy (which is not yet visible in the LEI) – or (ii) is simply wrong (and there will be no recovery in the LEI in the near future) – or (iii) S&P500 increased for other reasons which, by definition, cannot affect the LEI level. Bonus Chart. Yesterday we got the November 2023 flash reading of the S&P Global US Manufacturing PMI, which dropped to 49.4 (from 50.3 in October). Chart 3 shows the changes in the two main US PMIs and the annual change in the LEI. All three leading indicators confirm an upcoming slowdown in the US economy. Or all three indicators are wrong at the moment, and it’s just the S&P500 that correctly indicates the upcoming recovery in the US economy…
LEI vs Russell 2000
27-Nov-2023. Leading Economic Index vs Russell 2000. A quick comparison of the Leading Economic Index not only to the S&P500 but also to the Russell 2000. Figure 1 shows the annual change in the Russell 2000 and S&P500 based on the prices on the last trading day of the month. Figure 2 shows the annual change based on the average monthly prices (Figure 2 also shows the indexes themselves since 2006 - see bottom panel). Both stock indices have behaved similarly over the years, but it is clear that in recent months "gravity" has not spared small companies and the Russell 2000 is “lower and closer" to the LEI. This is even more visible when we take a closer look at the S&P500 and Russell 2000 (Figure 3), and especially the relative strength between them. Between February 2022 and March 2023, relative strength was roughly flat (both indexes moved similarly). However, since March 2023, the Russell 2000 index has definitely underperformed the S&P500 (which has also coincided with the lack of LEI recovery, which annual change since April 2023 has been flat at the level of -8% to -7% (no rebound here)). The outperformance of the S&P500 is mainly due to Mag7 (the generative AI, among others). However, it seems that the Russell 2000 and LEI are more indicative of the condition of the local economy.
ISM manufacturing
29-Nov-2023. ISM Manufacturing. Forecast for November 2023. The ISM Manufacturing PMI is highly correlated with the regional PMI indices (New York, Philadelphia, Dallas, Richmon, Chicago), which we used to get earlier in the month before the release of ISM Manufacturing itself. We already know the first 4 regional PMIs for November, the Chicago PMI will be announced tomorrow (consensus 45, up from 44 in October), and on Friday the ISM Manufacturing (consensus is 49, up from 46.7 in October). My very simple model shows the ISM rebounding to 47.5 (Figure 1). The average of the regional PMIs is currently up +0.097 of standard deviation (I calculate the average based on the 4 already known readings + consensus in the case of the Chicago PMI). Figure 2 shows the longer history of ISM Manufacturing and the Average Z-score of the 5 Regional PMIs (0.91 correlation with ISM Manufacturing). Figure 3 shows the contributions of each regional PMI to the Average.
Chicago PMI Nov-2023
30-Nov-2023. Chicago PMI. Just printed 55.8! Wall Street expected only 46 (was 44 in October, so for Chicago PMI, this is an increase of 1.44 standard deviations). This raises my forecast for ISM Manufacturing to 48.3 (from 47.5) – see the chart. It should be remembered that PMI indices are indexes of the width of change, not the size of change (in other words, it is enough that in "most entities" it increased by 0.01% and we will get a high reading). Some comments suggest that the Chicago PMI’s beat is due to the return of workers to factories after the recent strikes.
30-Nov-2023. Chicago PMI. Just printed 55.8! Part 2. You can see how much the Chicago PMI soared in the attached chart. The difference between ISM Manufacturing (October) and Chicago PMI (November) is just 1.29 standard deviations. However, such divergences have historically occurred, but very rarely (since 1990, in the same month, only three times has the Chicago PMI been more than 1.3 standard deviations higher than the ISM Manufacturing).
FedEx & UPS (Dec-2023)
21-Dec-2023. FedEx -12%. Are we closer to the next S&P500 peak? FedEx and UPS are bellwethers for world trade/economic growth. So their stock prices could have some leading properties… Before the S&P500 peak in February 2020, both companies, but especially FedEx, underperformed the S&P500 (see Figure 1). Before the S&P500 peak in January 2022, FedEx underperformed the S&P500. And today both are underperforming S&P500. UPS from April 24, 2023 (the publication of Q1 2023 earnings). And now FedEx has joined in, falling 12% yesterday after reporting earnings for the quarter ending November 30, 2023 … Well, are we closer to another peak on the S&P500?
New York PMI (Jan-2024)
17-Jan-2024. US New York Fed Manufacturing PMI for January dropped like a rock and amounted to -43.7. In December 2023, the New York PMI was -14.5, and the market expected an increase to -5.0 in January 2024. Figure 2 shows the volatility of the New York PMI, which is particularly high from 2022 onwards. A strong rebound can be expected in February 2024. Additionally, the New York PMI shows the lowest correlation with the ISM Manufacturing PMI among the other 4 regional PMIs (Dallas, Richmond, Chicago, Philadelphia).
FedEx & UPS (Jan-2024)
21-Jan-2024. Poor earnings and weak guidance were published by UPS and the price fell 8.2% yesterday. As a result of poor earnings, UPS also announced the layoff of 12,000 employees. Both UPS and FedEx are companies whose relative performance compared to the S&P500 can be treated as a good leading indicator. Both still indicate a slowdown in the economy... UPS underperformed the S&P500 from April 2023, and FedEx from October 2023 – see Figure 3.
PMIs & LEI (Jan-2024)
24-Jan-2024. Can we avoid a recession this cycle? In my opinion, this question is a bit misleading. A recession cannot be avoided, but it can be delayed. In this sense, we can avoid a recession in 2024. But if it is postponed, it may happen, for example, immediately after the US presidential elections, e.g. in Q1 2025. For instance, large spending in the form of a high fiscal deficit may keep the economy from entering a recession for many quarters, but at the end of the day the effect should be temporary, if only because deficit spending usually is not productive in aggregate and does not increase the wealth or prosperity of an average consumer. Where do these thoughts come from? We're just getting more data points for the most historically proven leading indicators. Yesterday we got the Richmond Fed PMI, on Monday the Leading Economic Index (LEI), previously the Philadelphia FED PMI and New York FED PMI. Today we will get the flash PMIs for January 2024. Figure 1 to 3 show the three regional PMIs. Figure 4 shows the LEI. But it seems that these indicators do not work in the current cycle because they have been pointing to a recession that never happened. So why don't they work? First of all, the legacy of the pandemic has not yet been fully behind us. Secondly, most of these indicators are based on the industrial part of the economy and do not reflect well the current strong services part. But that's why we call them leading, because it is the industrial/goods part of the economy (not the services one) that has historically led the cycle.
US Regional PMIs Jan-2024
30-Jan-2024. Regional PMIs. Yesterday we also got another US regional manufacturing PMI from the Dallas (Texas) region. This is another strong decline in the regional PMI in January (only the Philadelphia PMI did not decline - see Figure 2-5). This is starting to be disturbing... on Wednesday we will get the Chicago PMI, and on Thursday ISM Manufacturing. According to my simple model, assuming that the Chicago PMI does not change in January 2024 (it will be at the same level as in December 2023) - the model indicates an ISM Manufacturing reading of 44.9 – see Figure 6). This is quite low and these are levels that are difficult for the market to ignore (although PMIs in Europe have been lower for a long time, however the market certainly attaches greater importance to US ISM Manufacturing). My model has a 0.9495 correlation with the actual ISM reading. But it's just a model. In recent months, as can be seen in the Figure 6, we have had some decoupling between the model and the current ISM series than historically. Additionally, the flash reading for the S&P Global US Manufacturing PMI for January was quite positive: (Jan) 50.3 vs. Exp. 47.9 (Prev. 47.9).
Chicago PMI May-2024
1-Jun-2024. Abysmal Chicago PMI. On Friday, we also got the Chicago PMI for May, which dropped to 35.4 points! And these are recessionary levels. In general, levels below 35 points are associated with a deep and advanced recession - see Figure 1. After data normalization (Figure 2), the reading for May still remains significantly deviated from the ISM Manufacturing PMI, but it can be said that we are on the verge of the largest historical differences between the Chicago PMI and the ISM Manufacturing PMI. Data for April indicate 1.22 sigma, but if ISM Manufacturing had not changed in May (data next Monday), the difference would have been 1.52 sigma - and this would be the second largest ever difference in the same month (after June 2020). The Chicago PMI feels like an aberration if we also look at other similar series - see Figure 3. It should also not be forgotten that the PMI indices only measure the breadth of the change, not its size.
