Can NVIDIA Stock Rally Forever?

This article will serve to clarify some of the unclear points I have made in the past few articles. It’s important to analyze how you can be wrong to avoid sticking to bad bets when the tide begins to turn. For example, if you’re bearish on oil, but know which factors could push it higher, you will cover your shorts when you see the factors you’ve researched playing out. This can save you money. In terms of clarifications, sometimes charts can be made to show a point which isn’t true. Once you find out something you thought to be true isn’t true because another set of data provides a new perspective, it’s best to change your mind right away and not hold on to faulty arguments.

One aspect which I may have been unclear on is Chinese inflation. I have mentioned the Chinese inflation surprise index has been cratering. That index is only a somewhat related derivative of the actual data. The actual inflation, as you can see from the chart below has been rising in the past few years. It looks like the non-food CPI growth rate is stagnating, but it would be wrong to say that Chinese inflation is decelerating. The reason for the decline in the surprise index is because commodity prices are falling and because the estimates have been overzealous.

The problem with surprise indexes is they don’t tell you the direction of the data. You would think if the data is disappointing, the analysts will lower their forecasts so they don’t miss by as much, but there’s no rule that says that must happen. The Citi macro surprise index isn’t correlated with the stock market and is only tangentially related to the economic reports themselves. Surprise indexes are a tool to discover how reports are doing versus consensus, but they don’t give you a summary of the economy or whatever metric it’s covering. This line of thinking was true with the chart I showed in a different article which compared the earnings beat rate to the stock market. They weren’t correlated. To be fair, that was less related to the actual data than these surprise indexes because there could be a case where the firms which are missing expectations are missing by a lot, while the firms which are beating estimates are barely beating. That example scenario is actually fairly reflective of reality most times as analysts set the bar slightly too low so most firms can beat estimates by a smidge.

The next point I want to make is that some technology firms are dramatically expensive. Tesla, Snap, and Netflix are all trading at expensive valuations. It’s tough to value Tesla and Snap because they don’t have profits. The chart below attempts to value Tesla on its 2018 profit estimates which may or may not come true. Amazon had a relatively seamless transition from breaking even to growing its profits, but that was because of the high margin Amazon Web Services division. On the other hand, Tesla acquired the dead weight that is the SolarCity business. As you can see in the chart, Tesla is trading at over 10 times Ferrari’s price to earnings ratio despite Tesla having no track record of profits. Tesla has become the go to stock for those betting the ‘risk on’ trade will continue. The shareholder base is so dedicated to owning Tesla, the latest results don’t mater.

The point where I could have been clearer in previous posts is how expensive NVIDIA has gotten. NVIDIA had been the best performing stock in the S&P 500 in the past 12 months up until its recent correction. While NVIDIA doesn’t affect the Nasdaq as much as the big 5 tech names, it’s important to recognize that its parabolic move higher makes the tech sector riskier than I portrayed. As you can see, NVIDIA’s price to sales ratio has soared. The company must deliver on high estimates if it wants to maintain its price. Usually chip names go out of style quickly which makes this stock vulnerable in the intermediate term if results sour.

The biggest point which I have analyzed incorrectly in the past few articles is how I have mentioned the big 5 tech stocks are having an outsized impact on the stock market compared to historical precedent. While the big 5 tech stocks have driven the market higher since December and they are nearing $1 trillion market caps each, the effect on the S&P 500 is much smaller than in the 1990s and isn’t even high historically. As you can see from the bottom chart below, the top 5 stocks in the S&P 500 are adding 0.5% to returns which isn’t historically high. It was 1.5% in the early 2000s. Microsoft stock was up 28.2% in December 1999. That type of speculation can only be seen in the smaller names I’ve mentioned before like Tesla and NVIDIA. Snap stock has had a terrible time in the public market after doing its IPO. This shows there’s still some relationship between the strength of a business and the stock. While I think Apple stock is a long-term sell, the valuations haven’t gotten stretched like in the tech bubble.

As you can see in the chart below, the capitalization share the information technology sector has of the S&P 500 is the highest since the tech bubble. However, the earnings share is higher than ever before. If you think the tech sector is a sell, that’s because you disagree with the estimates on future profits. It’s much different from the late 1990s where buying the tech stocks was a bet on earnings appearing from businesses that didn’t have real ways of generating them.

The final chart below shows the long-term earnings growth consensus in the S&P 500 and the information technology sector. Earnings growth estimates have come down dramatically. I think that’s a good sign because expectations are more realistic. The one problem is that the American economy is more heavily indebted and growing slower than in 1999. However, that’s an economy wide problem which isn’t specific to the tech sector.

Spread the love

Comments are closed.