How to think about the ongoing correction? Is it a 10% correction or a big crash?
we're at the juncture
This article focuses on the S&P. Composition, valuations, technical indicators to watch.
Overall, it looks bad with a strong downward momentum. This market does start to expect a recession.
This is part 1. Part 2 on Monday: specific stocks
The Market Cap of S&P500 is US$47tn. My index replica of the S&P150 has a market cap of US$41tn, so I assume it’s a good proxi for SPX overall. For charts I use the SPY ETF.
My S&P150 index includes 150 stocks, 31 Tech stocks (I reclassify Amazon, Google, Meta and Tesla as Tech), 24 Financials, and all the rest.
Basic facts
Market cap
Market cap of Tech and Non-Tech is 50:50
Both Tech and Non-Tech have appreciated a lot over the past 10 years, but Tech has outperformed a lot. Over 2013 to today:
Tech up 10x
Non-Tech up 3x
Valuations
You often hear that the market is expensive, either S&P or Nasdaq. Here is the reason to separate Tech and Non-Tech:
Tech is very expensive at 30x trailing EPS (that’s 2024 EPS) and has re-rated constantly for 10 years (see doted regression line in chart below). Tech is at the top of its PEx range (2nd chart below)
Non-Tech is expensive at 20x trailing EPS (2024 EPS). Non-Tech has re-rated very modestly, but is also at the top of its PEx range (3rd chart below).
Growth
The reason for Tech outperformance and expensive valuations is higher growth. Over 2013 to 2024:
Tech revenue growth 3.5x net income growth 4.3x
Non-Tech revenue growth 1.6x net income growth 1.9x
Little corrections and big crashes
In 2025, S&P and Nasdaq have just corrected by -10%. At the moment we do not know if this will become a big crash. But we have signs.
A little correction and a big crash start from the same starting points:
Valuations are high
You’ve made money and you sell wanna sell some stocks to lock-in profits
You start to worry about something. Maybe growth will slow. Maybe interest rates, maybe a trade war, maybe AI is a fake, maybe something else.
The stock market is a weighting machine, it weights news and risks – and it likes to worry. Paul Samuelson: “the stock market has predicted nine out of the last five recessions”. Samuelson said that in 1982 and it remains true: the false positive are the 10% corrections or “news scare”. Throughout 2024, we’ve had a very persistent fear of recession: inflation remained too high, rates too high, the economy had to slow and enter a recession. For ex here. It didn’t happen.
A big crash starts with a little correction, and then fears materialize: there is indeed a recession (or some other shock to the system).
But very often, the big crash is triggered by an exogeneous event, not something that was expected / feared could happen: a virus, a (trade) war, a bankruptcy. Sometimes you just cannot pin down a clear reason. This wiki article on the dot-com crash is very interesting: why did it happen? No single factor can explain it. Soros calls it reflexivity – you got to read this book.
High valuations do not, by itself, start a correction. The exception is maybe the 2000 dot-com crash (since there isn’t any other clear catalyst) but as you can see below (same chart as above), valuations can stay elevated for a long time. The “valuation vigilante” and other thrift investors aka “value investors” will scream “stretched” and “expensive” all the way up. It does not help: valuations can stay elevated for a long time.
Finally: timing.
If there is a sudden exogeneous shock (Covid, Lehman bankruptcy, Greece bankruptcy), the stock market suddenly crashes with the event.
In case of an economic recession, the stock market has the uncanny ability to start its correction ~6 months before the recession is confirmed in the official data (NBER, BEA, FED). This makes investors’ life a lot more difficult because the stock market keeps correcting down, but the economic data still looks ok. This is when market commentator do what they do: make up stories that justify the market direction. Truth is: there is no data available to confirm an upcoming recession. What happen is that the “collective wisdom of the investing crowd” starts to feel cranky, has bad feelings, would rather sell a bit more, would rather not buy. We do care about the collective mood. It does move the market. For professional investors, i.e. fund managers, the risks are a-symmetrical: if your performance is great, you get paid a bit more. If you loose too much money, you get fired.
To visualize that the market starts correcting 3-6 months ahead of the factual recession, this is a good article.
Where do we go from here?
I mentioned that we do not know if the ongoing correction will become a big crash. But we have signs. They aren’t good.
I use the SPY ETF instead of the S&P500 Index for technical reasons that have to do with my data feed. It’s the same thing.
Bad sign 1: valuations (charts above) are at a level vulnerable to a large correction. Again high valuations alone do not trigger a correction.
Bad sign 2: broke down below 200-day moving average (red line), not sitting on the 300-day moving average (green line).
Investors who bought the S&P after Aug 2024 are now losing money. It’s a question of resistance to pain: if the S&P keeps going down, they will sell.
If the ongoing correction stops on the 300-day mva line, that’s good.
Bad sign 3: the index has broken below support levels
The next stop is -5% below
The next next stop is -14% below
Bad sign 4: daily moves. Up days (the dark candles), fail to reach the bottom of the previous days decline (red candles). The positive dark candles fail to go above the blue lines. In other words, the up-days are smaller than the down-days.
How will we know if it gets worse?
At the moment, I think the clearest resistance is the 300-day moving average. If the S&P and Nasdaq breaks down below that level, it’s bad.
S&P500 is at the 300-day mva: 5,510
Nasdaq is at the 300-day mva: 17,555
The second thing to watch is the momentum: if the next days up fails to go back above the previous gap down, it’s bad.
S&P500 previous gap down: 5,705
Nasdaq previous gap down: 17,845