Analyst Mark Hulbert recently raised the question of what the rate of initial public offerings means for the stock market. He looked at IPOs and dividends as a broad way to evaluate investor sentiment. He used a variety of different data points to gauge if stocks were in a bubble. He also drew on (among other things) some interesting data from Jay R. Ritter, a professor at the University of Florida, who is an expert on IPOs. This led to some interesting observations.
Consider these five issues, in order of persuasiveness:
No. 1. Volume: In the first quarter of 2000, there were 123 IPOs. In the first quarter of this year, there were 58 — less than half as many, according to data compiled by Ritter.
No. 2. First-day returns: The average return for an initial public offering in the first quarter of 2000 was a blistering 96 percent. This year, only 22 percent.
To put this into broader context, have a look at these first-day percentage increases from that era based on Ritter’s research:
Va Linux (12/09/99) 698 percent
Globe.com (11/13/98) 606 percent
Foundry Networks (9/28/99) 525 percent
Webmethods (2/11/00) 508 percent
Free Markets (12/10/99) 483 percent
Cobalt Networks (11/05/99) 482 percent
MarketWatch.com (1/15/99) 474 percent,
Akamai Technologies (10/29/99) 458 percent
Cacheflow (11/19/99) 427 percent
Sycamore Networks (10/22/99) 386 percent.
There’s nothing like this happening today.
No. 3. Dividend Premiums: Malcolm Baker and Jeffrey Wurgler published a paper in 2006 titled “Investor Sentiment in the Stock Market.” Baker is at the Harvard Business School and Wurgler is at New York University’s Stern School of Business. Both also do research for the National Bureau of Economic Research.
The study compares the relative valuations of dividend-paying companies with those of more speculative companies. Hulbert observes that:
As exuberance reaches extreme levels, investors become bored by established, dividend-paying companies. In March 2000, speculative companies on average had a 43 percent higher valuation than the dividend-paying stocks. The comparable premium today for stocks in the S&P 1500 index is 26 percent, according to data from FactSet.
The key difference here is that with rates so low, dividends have become an alternative to fixed income. That makes this data point a little less compelling than it might have been otherwise. Still, it points to a lower risk appetite.
No. 4. Equity Issuance: Companies raise cash from the secondary market when that window is open. Can that be used to evaluate periods of speculative excesses? Perhaps. In the first quarter of 2000, corporate cash derived from equity issuance was 20 percent; the most recent data is from the fourth quarter of 2013, and it showed the equity share of cash was only 11 percent.
Two caveats on this data point: First, using the fourth quarter versus the first runs into seasonal issues. Second, the bond market has become a very inexpensive way for companies raise cash, especially with the 10-year Treasury yielding less than 3 percent. Hence, this metric isn’t an apples-to-apples comparison.
No. 5. Share turnover: Comparing share turnover is another way to distinguish the bubble era from today. The turnover rate in listed shares on the New York Stock Exchange in the first quarter of 2000 was an annualized 89 percent. In the first quarter of this year it was 58 percent.
This is the least persuasive data point to me. Exchange traded funds and the move away from active trading has dramatically reduced trading volume. Indeed, almost the entire 180 percent rally since the March 2009 low has been on volume that is lower than normal.
Taken as a whole, these five points suggests that speculation hasn’t run rampant today the way it did in 2000. The list above contains two strong points, one moderate and two that perhaps could be rationalized away.
The conclusion is that we are not in a speculative bubble.