Recent media coverage has focused on the possibility of a bubble forming in the stocks of companies involved in developing and utilizing artificial intelligence. Although a few of these stocks have experienced substantial gains, including Nvidia, Microsoft, Alphabet (parent company of Google), Oracle, and Adobe, the fervent investment of Generative AI has yet to result in a bubble.
In order to understand this current situation, it is important to recall the elements of a bubble, as many market historians, economists, and myself have outlined. Notable examples include Charles Mackay's "Some Extraordinary Delusions and the Madness of Crowds", John Kenneth Galbraith's "The Great Crash, 1929", Edward Chancellor's "Devil Take the Hindmost" and Charles Kindleberger's "Manias, Panics and Crashes". These books provide insight into the tendency of investors to become overly enthusiastic about stocks.
The bubble books trace the story of the Dutch tulip mania in the 17th century, and the South Sea and Mississippi bubbles in England and France in the 18th century, as well as the stock craze of the Jazz Age in the Roaring '20s. Additionally, they chart the rise and fall of Japan's stock and property bubbles in the 1980s, the internet frenzy of the 1990s, and the most recent global real estate and credit bubble that ultimately lead to the Great Financial Crisis of 2008. There were several shared features in these bubbles, from an initial disbelief in the potential of a given asset or technology, to wider acceptance and eventual rapid increases in asset prices, culminating in a mass public participation and large-scale issuance of stock by connected firms.
We have all quickly come to understand the incredible potential of AI, however only a select number of companies have seen their stock prices rise in the expectation that AI will have a major effect on our lives and work. People have become more interested in buying tech stocks/ETFs related to AI but the entire stock market has not yet been taken by storm. This is evident when we consider there were 456 IPOs in 1999 amid the height of the internet mania with 77% of them having no profits. It was noted in the book "TrendWatching" that in 1998 and '99 around half of IPOs went up over 50% on their first day of trading and during 1999 1/4 of IPOs doubled on that same day. In 1999 the Nasdaq Composite rose by 85%, a record year for the index and by 2003 that had fallen by 75%. A bubble in AI might be on the horizon but at present it has not been caused by the Federal Reserve's "easy money" as it has not been put into AI stocks or any other asset class. The public is not yet fully invested so there is no bubble yet.
The gains in the Nasdaq 100, while impressive at 33% year to date, are concentrated in a handful of companies, much like the so-called "Nifty 50" of the 1970s rather than the internet bubble of the late 1990s. Some people go so far as to say it's impossible to pinpoint a bubble while it's growing, but I would argue that's not the case - the vast difference between a tiny bubble and a major one is easy to spot. A major bubble can crash entire markets and economies, like happened in Japan in the 1990s or in the U.S. after the real estate and credit crises. Artificial intelligence is certainly getting a lot of attention, as well as some investments, but it's not taking up all the available funds. Now may come the day when investors place bets on artificial intelligence without regard to actual earnings or profits and that's when smart money must be distinguished from dumb money. Commentary by Ron Insana, CNBC and MSNBC contributor and author of four books on Wall Street. Follow him on Twitter @rinsana.
Comments