Before we get into mid-term specific years let’s look at how the “average” year actually looks like. After all we want to be able to benchmark against what usually happens.

This chart is of every year from 1930-2025. Despite an “average” year returning 7.88%, the reality is that almost no years are in the 7-8% zone. Instead, we have a lot of years far higher, far lower, and ending up kind of flat. There is a definite upwards bias, but the “average” is rarely hit. This is worthy of its own post, but for now just realize that for scenario building, using the average is only a starting point.

We are breaking down the years with two different time periods. One is all the data we have for the SP500 from 1930-2025 and the other is the post-WW2 era. We like using all data, but a lot of people want to take out the Great Depression as they view it as an anomaly, and some look at post-WW2 as the modern era.  

Looking at 1930-2025 we can see that the average year delivers +7.88% per year. You can also see that for the most part the average is quite smooth with an average worst drawdown of only -1.52%.

If we look at the post-WW2 era, we get even better returns with +8.96% per year and an average worst drawdown of -1.06%. Funny how returns improve and risk drops when you take out the Great Depression.

With all that as our backdrop-average returns of +7.88% and +8.96% and average worst drawdowns of -1.56% and -1.06%-how do mid-term years stack up?

From 1930-2025 the average mid-term year delivers +2.83% per year with a worst drawdown of -4.95%. Worth noting is that the average mid-term year sees a lot of chop and is negative for close to half the year.

Looking at the post-WW2 era things improved a bit with the average mid-term year delivering +3.12% per year with a worst drawdown of -5.13%. So, your total return improves but your risk gets worse. Worth noting is that just like the full 1930-2025 period, the post-WW2 era mid-term year also sees close to half the year in a drawdown.

Since Trump is a 2^nd term President, we actually have one direct comparable with his mid-term year of 2018. As you can see in the chart below 2018 was all over the place. It went up +9.57% before dropping -20.02% and then ending the year down -6.23%.  Most of that year was in positive territory, but the back 70 trading days saw it all fall apart. That 70 days is also where the elections happen to occur.

What is interesting is that if you look at the 4-Year election cycle the 3^rd and 4^th years are usually strong as the incumbents do stimulus to try and juice the economy and help their election odds.  At the same time in 17 of the past 19 mid-terms the incumbents have lost the house and lost senate seats as well.  It is like the party might care about the Presidential election but doesn’t seem to care too much if the current President is able to do anything the second half of their term.

One more thing worth nothing is that on average after the elections the market tends to do quite well. Once the uncertainty clears, on average at least, it pays to be long.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

To conclude our series of The Most Important Chart In The World we have Chart #3, and it is a recycled chart from 2025. We didn’t want to do it, but we kind of had to.

Last year we showed a few different charts, but the first one was this one-SP500/MSCI World ex-US overlaid on the US Dollar-we used the Broad Trade Weighted USD Index, but you could use the DXY or the Bloomberg USD index and get a similar looking chart.

So why did we have to show this chart again? Well not only did it work out really well in 2025, but we suspect that this is just the beginning of a multi-year move.

In 2025 the SP500 did fine. It was up +17.9%. Great right? In a vacuum no one is going to be mad about that, but when you look at global markets it actually trailed. MSCI DM was up +21.6%, EM was up +34.4%, Japan was up +25.5%, UK FTSE was up +24%, MSCI Europe ex-UK was up +20.1%, etc. If you are noticing a trend, it is that global markets had a fantastic year.

Why did they beat out the US and why do we think this trend continues for a while, probably for years? Two main reasons. The first is simply valuations. US is very expensive vs the globe. The second is that US equities outperform global equities when the US Dollar is strong, and they trail global equities when the US Dollar is weak. And right now the USD is in a downtrend, and we suspect that continues for some time.

With that as the backdrop you can see why a chart like the one above or this one might make us think that being long global stocks is not only more interesting than it has been in years, but also that being overweight vs US equities could make a lot of sense.

Now before the doomers come saying “yep the dollar is crashing, and markets are going to lead us into the next great depression” let us be clear that this is a relative call and not an absolute one. US equities can continue higher in the face of a weakening US Dollar, it is just that they tend to trail global equities. At the same time a declining US Dollar doesn’t mean an outright crash, just that it has been strong against most major currencies for 15 years and for a host of reasons it is time for some mean reversion.

Conclusion

We think that all three of our “Most Important Charts In the World 2026” can be useful to help form some directional biases as well as longer term asset allocation decisions. While none of these are set in stone views-our views can change in the blink of a stop loss-they are at the top of our list of things to monitor into the new year.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

Continuing with our “The Most Important Charts In The World For 2026” series today we have global government bond yields

We watch yields all day every day but right now they are in an interesting place. US yields have been in a 140bps range since mid-2023 but the same can not be said for Government yields across the globe.

Looking at this chart of different government 10-Yr yields you might notice a trend. Japan, French, and German yields are in clear uptrends, and Aussie yields broke higher over the past few months. The one laggard? The US 10-Yr, which has been going mostly sideways for a while now.

If we clear it out and just look at the US 10-Yr you can see how it has been in a 140bps range for the past two and a half years. After a big move higher from the 2020 bottom, it has been rangebound since 2022.

If the US 10-Yr decides to turn up and join, or rejoin, global yields then we will likely see some fireworks in markets. On the other hand if the US 10-Yr goes lower and global yields follow we are likely to see another big bull run in risk assets.

Alternatively, and really what we hope for, is that they all kind of go in their own directions. This would help us to have a lot more less correlated trading ideas throughout the year. But for now the rest of the world is seeing mostly higher yields, and US yields are in a range.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

What has been the number one topic in markets over the past two years?

If you said AI then we are on the same page. AI and the associated power and semiconductor buildout have been the primary driver of equity markets as the hyperscalers are around  40% of the SP500 and it is rumored that OpenAI and SpaceX will go public in 2026.

As we covered in detail a few posts ago here-The SOX/SPX YoY A Semiconductor Cycle Leading Indicator , we have found one indicator that has been leading markets in the post global financial crisis era, and has done even better over the past 10 years as tech, and semiconductors, have been one of the biggest drivers of the overall market.

The indicator is the SOX Index divided by the SP500 and looked at year over year. The SOX/SPX Ratio YoY not only moves before the broader market, but it also does a great job of leading the Semiconductor Billings Index, which happens to move with the SP500.

Right now the SOX/SPX Ratio YoY as well as the Semiconductor Billings Index are towards their historic upper end. We are hesitant to say that as they are not off the charts, just getting into the normal upper part of the range. As you can see in the following chart of the Worldwide Semiconductor Billings Index YoY and the SOX/SPX Ratio YoY have definitely shot off the charts in the past, and there is nothing saying it can’t do it again.

With that we name this chart one of the most important charts in the world for 2026. It is a leading indicator for the semiconductor cycle, and as semiconductors are vital to AI expansion, and are also in almost every single electronic device we use today, this cycle has been, and is likely to continue to be, critical to catching the cyclical growth moves in the market.

If either the SOX/SPX Ratio YoY or the Semiconductor Billings Index YoY turns down, then we are likely to see equity markets drop. If on the other hand they keep moving higher and/or just go sideways, then markets are likely to continue moving higher. And before you run out and get super bearish keep in mind that they have gone sideways for months at a time before and can do so again.

Conclusion

A lot of things can happen, almost anything can happen, but if semiconductors continue to crush it then earnings are markets are likely to continue to do well. If on the other hand they fall on less than stellar times they are all likely to move lower. We are not in uncharted bubble levels, but we are in the neighborhood of a bubble.

You can create the SOX/SPX Ratio YoY with Excel or almost any charting platform, and you can get the monthly Semiconductor Billings Index data for free online. Based on the past 30+ years we think you would be wise to follow them both.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

Most Important Chart #1

First up we have SP500/MSCI All Country World Index Ex-US ratio with the US Dollar overlaid.

This was a win. Not only was it maybe the most actionable chart we posted, but all parts of it worked out. The US Dollar went down, and global stocks outperformed US stocks.

On a total return basis, the ACWI Ex-US is up +32.89% as of 12/29/25 and the SP500 is up +18.74%. Both went higher but foreign stocks beat the SP500 by +14.15%.  

The USD, and this depends on what USD Index you use, was down between -5 to -10%. We like the Broad Trade Weighted Index for general purposes, but you can use the DXY, the Bloomberg USD Index, etc. After the giant 15-year uptrend the USD got hit this year, and as of this writing is not exactly looking bullish.

Most Important Chart #2

The next post was about US equity valuations and forward expected returns.

This was a loss, or maybe a shove?

Valuations are a great tool, but as we even noted in the post, they are not a great timing tool. Instead, they can tell you a lot about the next 5-10 years, despite not providing high signal value for the next year.

We can claim a shove if you look at relative valuations and relative performance. US stocks were and are expensive, and US stocks underperformed foreign stocks.

We can call it a loss because US stocks still went up this year.

Go ahead and call it a loss, we should not have used that as a “Most Important Chart Of The Year” and instead maybe done a post on “The Next 10 Years” or something.

Valuations still matter and it will be surprising if the next 10 years brings anywhere close to the same returns as the last 10 years, but for now the trend is higher so what do you do?

Most Important Chart #3

Our last chart was of “Withheld Taxes-The Ultimate Coincidental Indicator” and it was a win.

Still an underrated data series, the “Daily Withheld Employment and Income Tax Balance” from the Daily Treasury Statement is a real time indicator of employment and wages.  The basic way to read it is to see if it is strong or weak on a year-over-year basis.

In our post we said that it was looking decent and that we expected it to continue to be that way for a while. Up until the last couple of weeks it literally flatlined YoY, and stocks kept going up. This despite the DOGE mess, continued layoffs in the corporate world, and the Government shutdown.  Turns out wage growth, even when it goes sideways, usually means that money is going into the stock market.

What’s Next?

Over the next few days we will have some new “The Most Important Chart In The World” posts. They will be easy to grade-as opposed to the valuation post-and all but one will be new.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

As with most sectors, especially those that manufacture physical goods, the semiconductor industry moves in cycles.

We can see this well if we look at the semiconductor revenues over time. In the chart below we have the Worldwide Semiconductor Billings Index from the World Semiconductor Trade Statistics. You can see the long-term trend (the blue line). By looking at revenues year over year (the red line), you can see the cyclical nature of semiconductor revenues. Some of the cycles are shallow, but most are quite large.

The next chart shows year over year data with recessions highlighted. With the data we have you can see that semiconductor revenues are weak ahead of each recession with COVID being an exception.

One thing worth pointing out, and this cuts across almost every cyclical tool, is that the good indicators indicate likely recessions…but they also indicate many non-recessions as well.  We have all heard the joke of “XYZ indicator or model called 10 of the last 5 recessions” as if that makes it a bad tool. It does not.

What we want are tools-call them models or indicators if you want-that show when a condition is happening. In the case of the semiconductor cycle we are looking for indications of growth picking up or growth slowing down. As you will see with the coming sections this is useful because it both helps us both to find recessions and, more importantly it helps us to find moves in markets. In our case that is the primary objective, signalling a recession is just a bonus.

ISM PMI

One of the more popular, and usually quite effective, leading and coincidental cyclical indicators is the ISM PMI-Institute of Supply Chain Management Purchasing Managers Index. The data goes back to 1948 and over time it has done a good job of showing us how the manufacturing industry is doing.

If we overlay the worldwide Semiconductor Billings Index year over year with the ISM PMI we find in this next chart that they tend to be very much in sync. Not only that but most of the time, especially over the past 25 years, we find that the Semiconductor Billings Index LEADS the PMI data. This has definitely been the case over the last few cycles as the Semiconductor Index leads up and down.

If the ISM PMI is a leading to coincident indicator, then this would imply that the Semiconductor Billings Index is a leading indicator of the cyclical turns in growth.

The current situation is interesting as the Semiconductor Billings Index is clearly showing a lot of strength while the PMI is not. In fact, in the last 36 months the PMI has only been above 50 twice. It would appear as though the only part of manufacturing showing strength is in semiconductors. We see the evidence of this in S&P sector returns as tech and communications have left the rest of the market in the dust.

Semiconductor Cycle and the Stock Market

So, we have established that the semiconductor cycle via the Semiconductor Billings Index is a useful tool to track manufacturing, growth, and by extension the business cycle, but what about markets?

Before we get there let us look at the PMI vs the SP500 year over year. As you can see in this chart the market usually tracks the PMI quite well. Sometimes the PMI leads, usually it moves right in line with the SP500. Of course as noted above relative to the Semiconductor Billings Index, and visible in this chart with the SP500 year over year, you can see that over the past three years there has been a big divergence, as overall manufacturing has been weak while the market has been strong.

This is solved, or at least in the current cycle it is, by looking at the Semiconductor Billings Index where we see a similar relationship where the Semiconductor Billings Index leads or coincide with most moves in the SP500. Worth noting is that in the current bull move the Semiconductor Index is right where it “should” be.

The Market Derived Semiconductor Cycle Leading Indicator

The Wisdom of Crowds is a concept that says when a collection of independent people makes a guess that they are usually more accurate than a single individual. The idea has been around since Francis Galton, but it was popularized when James Surowiecki wrote a book of the same name. It is important to note that this is not the same as crowd phycology as in the Madness of Crowds. The Madness of Crowds, or crowd psychology, is referring to herding behavior, while the Wisdom of Crowds is looking at predictions based on observations, or a giant collection of guesses.

A good example of the Wisdom of Crowds is if you have a jar of jelly beans and have 1,000 people guess how many jelly beans are in the jar. In studies using similar examples the median guess is usually very accurate. An example of Madness of Crowds on the other hand would be when you do something-for instance buy stock-not because of some rational reason, but because your friend bought that stock and is making money and you don’t want to be left behind. Both are very much at play in markets, but are distinctly different.

If we take the SP500 and look at it on a year over year basis, we can see that it has declined into each recession minus one in the post-WW2 era. The crowd of investors seem to be able to sniff out declining growth. Of course, you might also notice that like most cyclical tools the SP500 has called 10 of the last 5 recessions, so each decline does not mean it is a recession, but you don’t get recessions without a declining market. With this track record it should come as no surprise that this indicator is part of the Conference Board LEI-Leading Economic Index.

With that idea in mind what can we do to make a semiconductor cycle indicator using stock market data?

The first semiconductor stock index created was the Philly Semiconductor Stock Index-SOX- which has existed since the end of 1993. Since then, several semiconductor indices, mutual funds, and ETF’s have also been launched. But for the longest data series we have the SOX.

If we want to see how the SOX is doing versus the general market we can just look at the ratio or relative strength line. You do that by taking SOX and dividing it by the SP500. If the line is rising it means the SOX is stronger than the overall market, and if it is declining it means the market it outpacing the SOX.

In the following chart we show the SOX/SPX ratio over its history. As you can see when it is stronger than the market (blue line rising) the market is also almost always in an uptrend and vice versa. It is both a gauge of semiconductor stocks strength as well as a risk on/risk off indicator as the overall market, and not just semiconductor stocks, is stronger when the ratio is rising than when it is falling.

If we take the SOX/SPX ratio and then look at it year over year and overlay it on the Semiconductor Billings Index year over year, you can see that they usually move right with each other. In fact, the Wisdom of Crowds-the market version-actually leads almost every single move higher or lower as investors anticipate revenue growth acceleration before the ramp up or down in growth actually starts.

This is useful for a variety of reasons, but the primary two are that it is a true leading indicator of the semiconductor industry and as opposed to the other data we have looked at so far, it provides higher frequency data, which gives us “earlier” signals. Of course, that also leads to a bit more noise, but we like that trade.  

The SOX/SPX YoY Leading Indicator

Using daily data we can see how the SOX/SPX YoY performs against both the ISM PMI and the SP500.

In the first chart we have the SOX/SPX YoY and the ISM PMI. The semiconductor cycle leads almost every move in the PMI, and more importantly right now is that it is also catching the AI buildout while the overall manufacturing industry has been in a long slowdown. As the AI buildout and the internet of things continue to grow we can expect that the semiconductor cycle will only grow in importance in the coming years.

More importantly for our uses the SOX/SPX YoY leading indicator also leads the SP500 up and down more often than not. We can see that in the chart below that shows the two data series using daily data.

Simple Backtest of the SOX/SPX YoY Leading Indicator

To check the concept out we look at what would happen if you bought the SPX at the bottom of the semiconductor cycle and sold at the top each time. We also look at buying at the top and selling at the bottom.

As you can see in the chart below of each cycle since the data began, more than all of the positive returns from the SPX are gained when the SOX/SPX YoY indicator is moving higher from the bottom. When it is moving down from a peak we see slightly negative returns.

In the average bull cycle the SP500 went up 26.16% and in the average bear cycle it went up 0.68%. The best bull cycle saw it up 49.9% and the worst bull cycle, the only loser we have so far, was down -9.16%. The worst bear cycle on the other hand was down -39.8%.

This test is both simplistic and unrealistic, at least as is. How do you know it is “THE” bottom or “THE” top? That is for you to figure out. But if you look at other cycle tools as well as use some simple trend following tools you won’t get “THE” top or bottom, but you can get close enough to make this tool really useful.

If we look over its history, we can also look at the cycles themselves to try and approximate where it is likely to go both from peak to trough as well as by duration.

The average bull cycle lasted 446 trading days, and the average bear cycle lasted 495 trading days. At the extremes the longest and shortest bull cycles were 835 days and 165 days, and the longest and shortest bear cycles were 1042 and 340 days.

If we take the average peak and the average trough, we can see the average range to see what should be considered high or low. Combined with knowing how long the average cycle is, that is enough to give you an idea of how long and how far the cycle should go before it is likely to turn.

Looking at the average peak and trough, we get the chart below. We took out the top from 3/13/2000 because it was a HUGE outlier, and if anything, the cycles have moderated since then. But the purists might not like that, it is of course up to you. Since the dotcom crash, you might want to change the peak and trough lines as the cycles have seemingly moved into a tighter range, but we like it as is. As you can see it shows what would be considered the high and low end for a move. The closer you get to either line the higher the odds are that it will turn soon.

What The Semiconductor Cycle Says Right Now

As you may have noticed we are currently in a seemingly insane AI buildout. Datacenters are springing up like weeds, but these weeds use as much power as medium sized cities. And all we get are higher and higher revenue numbers out of the semiconductor manufacturers that only a few years ago would be considered all but impossible.

NVDA has a market cap of over $4.5 Trillion dollars and makes up over 8% of the SP500. In Q3 of this year it had over $51.2 Billion in revenue and they expect $65 Billion next quarter. Jensen Huang the CEO of Nvidia says he doesn’t see any AI bubble because the demand is so strong.

If you believe the bulls, we are only halfway on the trip to the moon. If you believe the bears we are in a deep recession and the AI bubble is about to pop.

You might not be surprised to find that we currently find things a bit in between the two camps. Yes, the AI buildout is in a bubble, but it could go on for some time. Yes, the SOX/SPX YoY indicator is close to the top of its range, but it still has space to go and not be an outlier.

Perhaps more importantly it is worth noting that we still think the semiconductor cycle is real and that despite the current situation we will see the cycle go up and down while the secular uptrend remains intact.

Remember this chart from earlier? The blue line-semiconductor revenue-just goes higher over time, but the line is not perfectly straight. Some time in the next 12 months it is likely to slow a bit, even if it keeps moving higher.

The Thing About Cycles

One last note about cycles. Use them to your advantage but do not get lost in them.  This tweet by Paul Enright is a great way to end this post.

Conclusion

There is no holy grail, but we think the SOX/SPX Ratio YoY is a worthwhile addition to your toolbox to gauge both the business cycle and the stock market. Try it. You make like it.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

Turns out they are a big deal, but only in the sense that they show the electorate how bad Congress is at their job.  If the shutdown lasts a few weeks, it also impacts employees and their families. But when it comes to markets they have so far been mostly a rounding error.

One thing worth nothing however is that the current one is just now hitting the all-time longest shutdown. Once you go past a month you have data collection issues, presumably different maintenance issues with equipment, etc. And we still don’t know when this will end. So saying “these are mostly a rounding error” does mean that they can’t become something bigger, just that so far they have not.

Before the 1980’s when the government had a funding gap the employees still went to work, or at most of them did. In 1981-82 the US AG Benjamin Civiletti called for more strict reading of the Antdeficiency Act. He decided that the government had no legal means to operate during a funding gap shutdown.

For instance, in 1980 under Carter there was a shutdown only of the FTC. That shutdown didn’t last a full day, but it happened. After the change of interpretation in 1982 this basically ended and now when it shuts down everyone is sent home…if there is time.

Another one, the February 9^th, 2018, shutdown only lasted nine hours so no one was actually sent home.

The November 1981 shutdown was a mess due to the recent change in the way the act was interpreted. The shutdown started Friday night, but since the employees needed to come in to start shutting stuff down, they still came in on Monday, but some were sent home at lunch. Of course it was all reopened the next day.

We went back through and collected data on shutdowns, their dates, and other info. We found it interesting that there does not seem to be one comprehensive place to find all the dates as it would appear the one day or less shutdowns don’t register the same with everyone, but they still happened.

With that we took all the dates and then looked at how markets did during the shutdown, for the three months after, the six months after, and the year after. We looked at the SP500, 10-Yr Yield, Gold, USD via the DXY, and Crude Oil.

We used closing data and if the shutdown ended on a weekend, we used the Monday close, if it started on a weekend, we did the same. Most start on Fridays, but over the years we have seen them start on every day of the week.

Here is the table for the SP500. We were not able to find much in the way of actionable data. President, party, who controlled Congress, etc. didn’t seem to matter too much for the stock market.

Next up we have the bond market. If the government is shutdown does it impact government 10-Yr Yields? Not really. The prevailing trends in yields and inflation carried on as if the shutdowns didn’t exist.

What about gold? Did people run for safety? Maybe they skipped bonds because the Government was toast? We see the same thing-in the late 70’s you wanted to be long gold, in the 80’s and 90’s you wanted to be short gold-but that was the prevailing trend and it showed no real edge during the shutdowns themselves.

The DXY-USD Index table does show one interesting thing, or at least it might. Most of the subsequent 3-12 months returns are positive. So does a rising USD enable lawmakers to not do their jobs? We will look into this, but based on the other asset classes my guess is that it is more chance than anything.

Finally we have Crude Oil. We used the front month of the WTI contract and didn’t have data going back before 1983 so the earlier years are blank. If you look at the aggregate stats vs the individual years you will see that aside from one outlier-2018-19-the returns during the shutdown are noise. Going farther out the oil table has the most mixed data of all of them. More noise.

Most of the shutdowns have not lasted an entire week. The ones that have so far have maxed out at 35 days. If you go and read about each one they are mostly just one party bickering with the other party over a specific part thing that they want, but it never seems to be THE thing that will end the republic. Consequently the markets seem to just look through them.

One day this could end as a shutdown lasts months and months-the current one is not over yet- or maybe it is about something that is the end of the republic, but until then we tend to think they are more noise than signal.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

Withheld Taxes-The Ultimate Coincidental Indicator

Everyday the Treasury puts out a statement showing the daily cash and debt operations of the U.S. Treasury. In this report one of the lines is the Daily Withheld Employment and Income Tax balance.

If people are getting paid more, or if more people are getting paid, or both, then this series heads higher. If people are getting paid less, if less people are getting paid, or both, then the series goes lower. So it shows a daily picture of wages and employment in almost real time. We say almost real time because the data is a bit messy on a day to day basis. Bi-weekly pay periods, end of month, end of quarter, etc. will do that.

If we look at this data on a year over year basis we have a great indicator showing earnings across a broad swath of the economy. In this chart we have the year over year data in blue, and then the 12-Month Rolling Sum in red.

As you can see it takes a strong recession for tax deposit growth to drop negative on a year over year basis. You need both higher unemployment and lower wages to take this one negative. Even in the bad times most people are taking home a pay check. Since 1983 we have only had three “normal” times where the year over year data went negative. We also had COVID and the post-COVID move where it went negative, wildly positive, and then negative again as the data normalized at the same time we had a bit of a slowdown.

What you will also notice is that we bottomed in February of 2024 and have been ramping higher ever since, crossing the 0% line in April 2024. Right now it is above 4%. In our view this indicates a fairly healthy employment and wage situation, which is another way of saying a fairly strong consumer.

So why do we think this is one of The Most Important Charts In The World right now? Well it is at 4% and usually doesn’t peak until the 8-11% area. Once it peaks it rarely just turns around and goes back down. Usually it oscillates for months and years above the 0% line. Go check out the SP500 year over year and see for yourself if that tracks.

We thought it was worthwhile to also show the rolling 12-Month sum in it’s own chart so you can see how it has acted the last few years. You can see the 2020 move lower, the giant move higher, and then the effects wearing off all of 2023 at the same time that we were seeing weakness in employment particularly in the tech sector. But you also can see how it ramped higher for almost all of 2024.

What Next?

Anything can happen, but it is not only rare, but outside of the the pandemic we have never seen the withheld taxes run up and then tank. So if the last 40 years are any guide we can likely expect wages and employment to be decent, if not perfect, moving forward for a while.

Yes, we have a new President coming in and depending on your political death cult you might think that this is the best or the worst thing ever, but based on history it is probably neither. Under the last seven Presidents we have not had one huge downturn where you can lay all the blame on one party let alone one President.

Looking at the data it is clear that we are blessed to have an economic juggernaut that usually just plows forward regardless of all the noise. This is a feature and not a bug.

Anyways we expect withheld Federal taxes to stay strong for some time moving forward. You can choose for yourself what this means regarding your positioning.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

US Equity Valuations And Forward Expected Returns

There are several models that attempt to forecast future stock market returns. Most of the time they are based on valuations.

It should come as no surprised that if you invest at low valuations there is more upside than if you pay at the upper end of valuations. Paying single digit price to earnings will usually do a lot better than paying 30 or 50 times earnings.

If we look at maybe the most famous of the long term SP500 models we have the Shiller CAPE Ratio. This ratio takes the average earnings of the past 10-Years and then divides it by price. It is based on an idea from Ben Graham to smooth out the earnings over a cycle.

Over time it has a great track record. If you buy when it is low then you do a lot better than if you buy when it is high. As you can see in the chart below the current readings are some of the richest of the past 135 years.

So should you sell and go to cash? Well that is where these models can bite you. This chart is done using monthly data going back to 1881. What you might notice is that all of the major overshoots lasted YEARS before coming back in. If you had sold in 1996 when it broke out of its historical range you would have lost another 4-5 years of great returns. In fact the ensuing recession only took the SP500 back to 1997 on a total return basis.

If you had sold the next time it broke out of the historical range you would have sold stocks in 2014….ten years ago. In case you have been on a deserted island the last 10 years the SP500 is up 300% since then.

Relative Returns and Risk Premiums

Another way to look at long term valuations and future expected returns are by using risk premia models. Maybe the most used model is made by taking the forward earnings yield and subtracting the 10-Yr Treasury Yield.

If you do that then you get the chart below. It is better to be buying stocks when it is high than it is when it is low. And right now it is negative as the 10-Yr yield is higher than the forward earnings yield on the SP500.

If you run a scatter plot of 3, 5, and 10 year returns you will find that there is indeed an edge in buying when the spread is high vs when it is low. But as you look at that chart and the date axis again you might notice something. In the 90’s it went up and down but the levels were not comparable to the last 25 years when up until the last two months the spread was always saying be long stocks.

There are lots of ways to tweak the model to make it useful. Something like Bollinger Bands or a regression model can work, but just going off the chart shows that it is not just a clean “buy here and sell here”…not at all.

Investor Positioning

There are a lot of ways to look at investor positioning, but most of them look at shorter term data like futures COT data, mutual and ETF fund flows, and options metrics. But these are more for tactical trading and not long term asset allocation. It turns out we have some great long term data.

For long term metrics we go to the Feds Flow of Funds report and get the data via FRED. The equation to show the Average Investor Allocation to Equities is the following.

Investor Allocation to Stocks (Average) = Market Value of All Stocks / (Market Value of All Stocks + Total Liabilities of All Real Economic Borrowers)

In the chart below the blue line is the allocation and the red line is the realized 10-Yr total return of the SP500 from that date. So the red line is always lagged 10 years back.

As you can see the relationship has historically been very good. In fact up until now it might very well be “The Greatest Predictor Of Future Stock Market Returns” , at least that is what Jesse Livermore from Twitter called it when he wrote the piece that brought this model to light.

So what is it saying right now? It is pointing to negative total returns over the next 10 years from the SP500. But despite the relatively tight fit there is an issue with this model as well. 10-Years is a long time, and most investors cant sit through 2-4 years of underperformance because a model was very good but not perfect. Also as with all models “what if this time is different?”

Problems And Solutions

We don’t have a holy grail, we wouldn’t tell you if we did, but we do have some ideas based on research, experience, and intuition that could help you make these models useful to you.

For the CAPE model, or any model you can duplicate for several different stock markets, you can make better use of them by buying the cheapest markets than by using the model to time just one. Meb Faber wrote a paper on this idea and has incorporated it into some of his ETFs.

Another tool you can implement is to pick a long term trend indicator that matches up with how you look at the world. For a lot of people long term would be the 200-Day moving average, for others it would be a lot longer or shorter. Figure out what works for you and overlay it with valuations. When valuations are at the upper end of reasonable maybe just stick with the trend until it ends, then change your allocation. Most of you, even large fund managers, can afford to give up a few percent at the top of bottom but catching the middle by using trend tools. Oh and you can and should back test this until you find a good compromise.

Use valuation tools as a rough guide, not as a fixed rule. If you are an active stock picker then pay attention to where we are in the valuations cycle but keep doing your thing. At some point you will see things stop working and you can start to move into different types of stocks. Obviously an expensive market can get not just more expensive, but insanely more expensive. We could write a book on this paragraph, and other people have, but being aware of forward expectations can be helpful to almost any investor.

Anyways as the title in this piece says these are some of the most important charts in the world right now and you will be seeing them a LOT over the next few months. In aggregate US stocks are not cheap right now, but this time is always a bit different so as they say “you do you”.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart

The chart is the USD overlaid with the SP500/Global ex-US Index Ratio. You can use the DXY, the BBG USD Index, or the Trade Weighted USD Broad Dollar Index. We find it tracks the latter the best, but they all say the same thing. Then you overlay it with the SP500 and World ex-USA Index ratio.

When the blue line is moving up it means US stocks represented by the SP500 are outperforming the rest of the world, and when it is going down the SP500 is underperforming the rest of the world.

The chart shows how the USD effects stock prices as well as how stock prices effect the USD. Since 1995 this relationship has been fairly strong only breaking down for short periods. Sometimes the USD leads and sometimes the equity ratio leads, but for each large swing they moved in the same direction.

This chart kind of goes against some of the research on the USD and its effects on US stocks. The research shows that a strong dollar hurts earnings for US corporations. And yet you can see that when the USD is strong, US stocks outperform global stocks.

One place where this chart and relationship should continue to shine is in asset allocation. If you are expecting the USD to turn then you probably want to overweight global stocks, if you expect it to continue higher then you probably want US stocks. There are several other takeaways, but this is the obvious one.

Worth noting is that prior to 1995 this relationship was not consistent. But with 30 years of data now, and with the relationship currently as close as ever, this is one of our nominations for the current “Most Important Chart In The World” right now.

Happy Trading,

Dave@PDMacro.com

P.S. If you liked this then take a free two week trial of our service. If you have any questions send me an email or find me over at Twitter @DavidTaggart