A Bit of Friday Excitement

Those who were hoping for a placid summer Friday were in for a rude surprise this morning.  Global markets, and thus pre-market index futures, were sharply lower as tech stocks took another leg lower.  The ostensible reasons were renewed Persian Gulf jitters that pushed oil prices higher, while reports that a new, inexpensive Chinese AI model rekindled DeepSeek-like concerns about that industry’s economics.  Throw in a reminder that monthly options expirations can still affect those mornings’ openings and we got off to a very sloppy start.

Let’s take care of the easier explanations first.  The situation in the Gulf has been deteriorating for the past few weeks, and while it has weighed on sentiment, it has not been in the forefront of the market’s mindset.  Crude oil futures are not yet back to the levels that prevailed during the height of the conflict, but they have been moving steadily higher since the start of the month.  It is difficult for stocks to entirely shrug off a roughly 20% bump in just two weeks in a crucial commodity.

3-Months, September Rolling Futures, CL (red/green candles), COIL (blue line)

Source: Interactive Brokers

The recent dip in oil prices led to an improvement in today’s University of Michigan’s preliminary consumer survey for July.  Sentiment rose from 49.5 to 54.9, well above the consensus 51, while 1-Year Inflation Expectations fell to 4.2% from 4.6%, below the consensus 4.4%.  Once again, we see the link between that inflation estimate and the price of gasoline that has generally held for years (with the notable exception of the post-Liberation Day tariff scare).

Since January 2010, UMich 1-Year Inflation Expectations (yellow) vs. AAA Regular Unleaded Gasoline (green) and CPI (magenta)

Source: Bloomberg

As for options expiration, this wasn’t a major factor in today’s move, but it contributed to the early angst.  There were nearly 60,000 contracts on the S&P 500 (SPX) with a 7,500 strike.  Once the futures were indicating an open below that level, some additional hedging-related selling emerged.  That contributed to stocks opening at their lowest levels of the day, and that is why we saw a reflexive bounce shortly thereafter.

But of course, the biggest issue today is the concern about cheaper Chinese alternatives to currently popular models from Anthropic, OpenAI, Alphabet, and others.  While I have no way to independently verify the veracity of Moonshot’s claims, the broad concern is that less expensive alternatives imply less need for spending on the technology necessary to build out more expensive models.  Hence, we see an already nervous tech sector pushing broadly lower. 

This brings back some unpleasant reminders. 

First, parabolic moves, like the one we had in the Philadelphia Semiconductor Index (SOX), are inherently unstable.  They often go on for longer than one might expect, but when they end, it’s usually unpredictable and painful.  That’s where we are now.

Second, the maker/taker theme that we discussed not long ago seemed to be switching back from the “makers” (semis, et al.) to the “takers” (hyperscalers), but that rotation petered out yesterday and is nowhere to be found this morning.

Finally, although VIX is higher today, it’s still at relatively low levels.  I’ve been asserting that the placid surface is hiding treacherous undercurrents.  Remember, VIX is not a fear gauge, but it plays one on TV.  There is nothing in its calculation that explicitly measures sentiment.  Also, index volatility depends heavily on the correlation and/or dispersion of its components.  High correlations imply higher index volatility, while high dispersion implies lower index volatility.  That is what we have seen recently with COR1M near 2-year lows and DSPX near a 5-year high.  Hence, we have a relatively low level of VIX.