Friday, February 7th, 2020

Chairman Mao is Coming to Dinner

In the introduction to The Moon’s a Balloon, his highly entertaining showbiz autobiography, David Niven says that “there is little point in writing about the butler, when Chairman Mao is sitting down to dinner.” In the same spirit, why bother with sector strategy or emerging markets, when the most important question is whether US equities, and particularly the Tech sector can continue its remarkable progress. Apple has a market cap larger than Germany. Five stocks in the US are effectively equal to the whole of the Eurozone and larger than Japan or the UK. This poses all sorts of questions about growth, valuation, regulation and geo-politics, which are effectively unanswerable. So, we are not going to try.

Instead of analysing the companies or their industries, we are going to treat them purely as securities. According to our standard model, the probability of Apple beating the return of the US index on a risk-adjusted basis (PRATER) is 100%. For Microsoft, the equivalent number is 98%. Even their biggest fans would agree that that this is not a situation which can last for ever. These scores are higher than they were in the most recent tech-boom in late 2017. However, our standard model is not equipped to give an opinion on when they might deteriorate, which is why we developed a process which can measure acceleration as well.

The new approach compares acceleration (in the form of a weekly RSI) with probability (which is equivalent to speed) and tries to identify which stocks are accelerating too hard, given their speed / position in the road. The big offender is Microsoft, with an RSI which is 2.1 standard deviations (STD) above where it should be. Apple is only 0.9 STD above its predicted value, having peaked at 2.3 STD two weeks ago. Both RSI scores peaked at 84%, well into sell territory, which conventionally starts at 70%.

Nothing in these numbers requires the stocks to start underperforming immediately. But we can look at other large caps and see what happened when they got to these levels. In August 2018, Amazon got to a PRATER of 100% and RSI 1.1 STD higher than its predicted value. Over the next three months, it underperformed by 18% and is still 12% below that relative peak, which looks like being the record high for some time to come. Neither Facebook or Alphabet have achieved a PRATER or an RSI score like this in the last three years, and it may not be fair to use evidence generated at a different stage of this long bull market.

In the last two years, we have been able to find only seven other large cap stocks in the US, which have simultaneously achieved a PRATER of more than 99%, an RSI over 80% and more than 1.0 STD above its predicted value. They are AMD in September 2018, American Tower in May 2019, Eli Lilly and Merck in December 2018, Netflix in June 2018, Starbucks in August 2019 and Xilinx in March 2019. (As of next week, Tesla will almost certainly be included.) Apart from American Tower, every one of them declined by at least 15% on a relative basis over the next 2-3 months. American Tower declined by 11% and is the only stock to generate a prior false positive, which would have cost a short-seller 9%. The typical maximum drawdown was between 20-40% relative to the index, occurring in the next 3-9 months.

This exercise suggests it is highly likely that Apple and Microsoft will start to underperform US Equities in the near future and if that’s right, US equities themselves will probably fall.

Leave a Reply

Full Blog Archive:

Other Recent Posts