Friday, March 20, 2015

Illiquidity and Bubbles in Private Share Markets: Testing Mark Cuban's thesis!

It looks like Alibaba is investing $200 million in Snapchat, translating (at least according to deal watchers) into a value of $15 billion for Snapchat,  a mind-boggling number for a company that has been struggling to find ways to convert its popularity with some users (like my daughter) into revenues. While we can debate whether extrapolating from a small VC investment to a total value for a company make sense, there are two trends that are incontestable. The first is that estimated values have been climbing at exponential rates for companies like Uber, Airbnb and Snapchat. In venture capital lingo, the number of unicorns is climbing to the point where the name (which suggests unique or unusual) no longer fits. The second is that these companies seem to be in no hurry to go public, leaving the trading in the private sharemarket space. These rising valuations in private markets led Mark Cuban to declare last week that this "tech bubble" was worse (and will end much more badly) than the last one (with dot-com stocks). In the article, Cuban makes four assertions: (1) There is a tech bubble; (2) A large portion of the tech bubble is in the private share market which is less liquid than the public markets; (3) The bubble will be larger and burst more violently because of the absence of liquidity; and (4) This bubble is worse than the dot-com bubble, though it not clear on what dimension and from whose perspective. In his trademark fashion, Cuban ends his article with a provocative questions,  "If stock in a company is worth what somebody will pay for it, what is the stock of a company worth when there is no place to sell it ?" I like Mark Cuban but I think that he is wrong on all four counts.

This is not a tech bubble
In my last post, I took issue with the widespread view that the rise in stock prices from the depths of 2008 has been largely due to tech companies using a simple statistic: the proportion of overall equity market capitalization in the United States coming from tech stocks. Unlike the 1990s, when tech companies climbed from single digits in 1990 to almost 30% of the overall market capitalization by the end of 1999, tech stocks collectively have stayed at about 20% of the overall market.

Tech stocks in S&P 500
There are other indicators that also support the argument that this is not a tech bubble, since a bubble occurs when market prices disconnect from fundamentals. Unlike the 1990s, the market capitalization of technology companies in 2014 is backed up by operating numbers that are commensurate with value. In the figure below, I compare tech companies to non-tech companies on market values (enterprise and equity) as well as on operating statistics such as revenues, EBITDAR&D, EBITDA, operating income and net income, across the entire US market (not just the S&P 500):
Tech vs Non-tech companies in US market (Source: Cap IQ)
One measure of whether a sector is in a bubble is if it accounts for a much larger share of overall market value than it delivers in revenues, earnings and cash flows. In February 2015, tech companies account for about 13.84% of overall enterprise value and 19.94% of market capitalization and they hold their own on almost every operating metric. While tech companies generate only 11% of overall revenues, they account for 19.99% of EBITDA+R&D, 17.93% of operating income and 16.46% of EBITDA, all much higher than tech's 13.84% share of enterprise value, and 18.65% of net income, close to the 19.94% of overall market capitalization. On the cash flow measure, tech firms account for almost 29% of all cash flows (dividends and buybacks) returned to investors, much higher than their share of market capitalization. To provide a contrast, in 1999, at the peak of the dot-com bubble, tech firms accounted for 30% of overall market capitalization but delivered less than 10% of net income and dividends & buybacks. That was a bubble!

Note, though, that this is not an argument against a market bubble but one specifically against a collective tech bubble. If you believe that there is a bubble (and there are reasonable people who do), it is either a market-wide bubble or one in a specific segment of the tech sector, say baby tech or young tech. In my earlier post, I broke tech companies by age and noted that young tech companies are richly priced. If Cuban's assertion is that young tech companies are being over priced, relative to fundamentals and potential earnings/cash flows, it is a more defensible one, and if it is just about young tech companies in the private share market, it may even be a likely one. Even on that front, though, the question remains whether this over pricing is a tech phenomenon or a young company phenomenon.

Illiquidity is a continuum 
Cuban's second point is that this bubble, unlike the one in the nineties, is developing in private share markets, where venture capitalists, institutional investors and private wealth funds buy stakes of private businesses and that these private share markets are less liquid than publicly traded companies. While the notion that public markets are more liquid than private ones is widely held and generally true, illiquidity is a continuum and not all private markets are illiquid and not all publicly traded stocks are liquid. 

The private share market has made strides in the last decade in terms of liquidity. NASDAQ's private market allows wealthy investors to buy and sell positions in privately held businesses and there are other ventures like SecondMarket and Sharespost that allow for some liquidity in these markets. To those who would argue that this liquidity is skin deep and will disappear in the face of a market meltdown, you are probably right, but then again, what makes you believe that public markets are any different? While it is true that some of the big names in technology have high trading volume and deep liquidity, many of the smaller technology companies often have two strikes against them when it comes to liquidity:
  1. Low Float: The proportion of the shares in these companies that are traded is only a small proportion of the overall shares in the company. Just to illustrate, only 10.5% of the shares in Box, the latest technology listing, are traded in the market and small swings in mood in this market can translate into big price changes. Looking across all stocks in the market, the notion that young tech companies tend to have lower floats is backed up by the data:
    Source: S&P Capital IQ (February 2015 data)
  2. Here today, forgotten tomorrow: The young tech space is crowded, and holding investor attention is difficult. Consequently, while many young tech companies go public to high trading volume, that volume drops off in the weeks following as new entrants draw attention to themselves, as evidenced by the trading activity on Box:
    Box: Stock Price & Volume (Yahoo! Finance)
The bottom line is a simple one. The liquidity in tech companies in public markets is uneven and fragile, with heavy trading in high profile stocks, in good times, and around earnings reports masking lack of liquidity, especially when you need it the most.  While Mark Cuban worries about the illiquidity of the private share market, I am not sure that it is any more illiquid than the public markets in dot-com stocks were in the 2000, as the market collapsed.

Liquidity can feed bubbles
Let us, for purposes of argument, accept that Mark Cuban was talking about baby tech companies in the private share market and that he is right about the private share market being less liquid than public markets, is he right in his contention that bubbles get bigger and burst more violently in less liquid markets? Intuitively, his contention makes sense. With start-ups and very young companies, it is a pricing game, not a value game, and that price is set by mood and momentum, rather than fundamentals (cash flows, growth or risk). If you cannot easily trade an asset, it would seem logical to assume that any shift in mood or momentum in this market will be accentuated. If you bring them together in a private share market, you should have the ingredients for a bigger bubble, right?

My intuition leads me down the same path, but if there is a lesson that I have learned from behavioral finance, it is that your intuition is not always right. Some of the most interesting research on bubbles, on what allows them to form, and causes them to burst, comes from experimental economics. Vernon Smith, who won a Nobel Prize in Economics for his role in developing the field, has run a series of experiments where he illustrates that adding liquidity to a market makes bubbles bigger, not smaller. To illustrate, he (with two co-authors) ran a laboratory market, where participants traded a very simple asset (that paid out an expected cash flow of 24 cents every period for 15 periods, giving it a fair value of $3.60 at the start of the trading, dropping by 24 cents each period).  Not only did they find bubbles forming in this market, where the price increased to well above the fair value in the intermediate periods, but that these bubbles were bigger and lasted longer, when they gave traders more money (liquidity) to trade in the market:


In addition, they found that adding liquidity made the bubble bigger earlier in the game. (I strongly recommend this paper to anyone interested in bubbles, because they also explored the effects of adding price limits (like futures markets do), short sales restrictions and experience.) Extrapolating from one experimental study may be dangerous, but if this study holds true, the fact that the private share market is less liquid than a public market may be a check on the market's exuberance, and especially so for young start-ups. Put differently, if liquidity adds to bubbles, Uber, Airbnb and Snapchat would be trading at even higher prices in a public market than they are in the private share markets today.

If you are struggling with the question of why liquidity adds to market bubbles, let me offer one possible explanation. A market bubble needs a propagating mechanism, a process by which new investors are attracted into the market to keep the price momentum going (on the way up) and existing investors are induced to flee (on the way down). In a public market, the most effective propagating mechanism is an observable market price, as increases in the price draw investors in and price declines chase them out. If you add, to this phenomenon, the ease with which we can monitor market prices on our online devices (rather than wait until the next morning or call our brokers, as we had to, a few decades ago) and access to financial news channels (CNBC, Bloomberg and Fox Business News, to name just the US channels) which expound and analyze these price changes, it is no surprise to me that bubbles have steeper upsides and downsides today than they used to. In a private market, we hear about Uber, Airbnb and Snapchat's valuations only when venture capitalists invest in them and our inability to trade on these valuations may be a restraint on their rising. 

A big bubble is not necessarily a bad one
The final component of Mark Cuban's thesis (though I believe that the first three are flawed) is that this bubble is "worse" than prior bubbles. But what is it that makes one bubble worse than another? To me, the cost of a bubble is not whether those invested in the bubble lose money but whether others who are not invested in the bubble are forced to bear some costs when the bubble bursts. It is that spillover effect on other players that we loosely call systemic risk and it is the magnitude of these systemic costs which made the 2007-08 banking bubble so costly.

With this framework in mind, is this young (baby) tech bubble more dangerous than the one in the late nineties? I don't see why. If the bubble bursts, the immediate losers are the wealthy investors (VCs, private equity investors, and private banking clients) who partake in the private share market. Not only can they afford the losses, but perhaps they need a sobering reminder of why they should not let their greed get ahead of their common sense. In a public market collapse, there will be far more small investors who are hurt, and though they deserve the same wake-up call as wealthier investors, they may less equipped to deal with the losses. This could change if institutions that have no business playing in the private share market (like university endowments and public pension funds) decide to invest big amounts in it and screw it up big time.

It is true that there will be side costs, as there are in any bubble. First, when a bubble bursts, the lenders/banks that lent money to companies in the bubble will feel the pain (which does not bother me) and then pass it on to taxpayers (which does). Since young tech companies are lightly levered, these costs are likely to be small.  Second, the bursting of a bubble can have consequences for governments that collect tax revenues from these companies (corporate tax), their employees  (income tax) and investors (dividends & capital gains taxes). Again, since young tech companies are money losers, the vast majority of employees settle for deferred compensation and investors in private markets don't cash out quickly, the tax revenue loss will be contained. Third, every burst bubble carries consequences for the real estate in the region (of the bubble). So, yes, the Bay Area will see a drop in real estate value, and is that a bad thing? I don't think so, since anyone in that area, who is not part of the tech boom, has been reduced to living in cardboard boxes. Finally, I believe that the collapse in the private share market, if it happens, will follow a collapse of young tech companies in the public markets (Facebook, Twitter, Box, Linkedin et al.), which I will take as an indication that it is public markets that lead the bubble, not private markets. 

If this is a bubble, I don't see why its bursting is any more consequential or painful than the implosion of the dot-com bubble. There will undoubtedly be books written by people who claimed to see it coming (perhaps Mark Cuban is vying for a front spot), warnings from the Merchants of Doom (you know who they are) pointing out that this is what happens when greed runs its course and there will be government/market/regulatory action (almost all of it bad, and most of it ineffective) to stop something like this from happening again. So, don't be surprised to see curbs on private share markets or on institutions investing in these markets, as if those curbs will stop the next bubble from occurring. 

Bottom line
Mark Cuban's entry into the ranks of the very rich was greased by the 1990s dot-com boom where he built a business of little value, but sold at the right time . Since that is how you win at the pricing game, I tip my hat to him. For him to point fingers at other people who are playing exactly the same game and accuse them of greed and short-sightedness takes a lot of chutzpah. In fact, Cuban's assertion about this being a worse bubble than the dot-com bubble gives us some insight into one very self-serving way to classify bubbles into good and bad ones. A good bubble is one where you are making money of the excesses and a bad one is one where other people are making money (or more money than you are) from the over pricing. If Cuban is serious about staying out of bubbles, he should look at the largest investment in his portfolio, which is in a market where prices have soared, good sense has been abandoned and there is very little liquidity. In a market where the Los Angeles Clippers are priced at $2 billion and the Atlanta Hawks could fetch a billion, the Dallas Mavericks should go for more, right?

9 comments:

Anonymous said...

It is consistent with the views of private equities which look at public market and IPO as a cash register at casino. In other words, if private market valuation is more expensive than public, the later stage funding would see unfavorable risk/return and it may push the later stage PE/VC toward an earlier stage and would make the period as a private company shorter. I see the later stage funding activity is still doing fine – a proof that public market (thanks to liquidity) can push private valuation upwards.
Bifurcation of tech companies seems to be in place – some public equity investors are fine with high multiple for early stage (tech & biotech) companies with negative stock comp adjusted FCF margin. There seems to be a segment in the market where equity investment and valuation are done “option”-wise like an investment in a new oil well. I do not think there is a broad bubbly expensive valuation for tech, but if there is any, that is something very common - a historically low discount rate because of QE.

DM said...

Thank you for your posts, I really enjoy reading them. My question is whether this is a bubble or an extreme over valuation (semantics?) of equities. As interest rates and qe have led to so much money being in the economy but no signs of low unemployment, wage growth or inflation present I wonder if that is bc what the fed uses to gauge inflation does not include equities (public and private)? Unless the current interest rate policy is the new normal, which I doubt, then it would stand to reason that the money HAS to go somewhere. And with the crazy stock prices in public (tesla, apple, twitter) and private (uber, snapchat) it just seems natural to think it is a bubble as this is where the money is going? Thanks again for your posts and insight.
DM

Katelyn said...

In response to Anonymous, from what I’ve read PE/VCs are on the contrary being pushed to later stage, given the rise of angel investors, seed funds, etc..

It’d be interesting to see if applying the same analysis to Biotech holds true as well, which is also seeing a PE/VC move towards later-stage activity. Bruce Booth did an interesting analysis (http://lifescivc.com) on this as well, that pulls some spot correlations between clinical and stock performance, as well as market cap and performance – would be interested to hear your thoughts.

Thanks for the great read!

Anonymous said...

In the trading side of the business, we use the term, "talking your book" as describing someone's bullish or bearish opinion to their actual positions in the market. I wonder whether Cuban is negotiating to purchase a company in the Tech space and using his public platform to reinforce his bargaining posture. The super-rich like Cuban operate on a different plane from the rest of us, so this would not surprise me.

Jim said...

Reading your article, I came to the same conclusion that it is public markets that lead the bubble, not private markets. For example, it's not surprised that more cash led to higher prices when traders were trading a single asset. That's no different than what has happened with QE and the public markets. Once a market starts to get saturated with money then the money will seek other markets. In this case, it is the private markets. Hence, public markets lead bubbles, not private markets, assuming that public markets take first investment priority because of their liquidity.

ABHIMANYU DHAMIJA said...

I think you should exclude Apple from your analysis.
The analysis should be Performance of S&P 500 vs Tech companies excluding Apple.

Apple seems to be undervalued and with Such high-revenues has to skew the tech sector overall.

Patrick Mathieson said...

It seems at least slightly incongruous that you're using public market metrics (of which, super healthy companies like Apple and Google will dominate from a market cap perspective) to disprove the bubble hypothesis as it relates to "private midcaps" like Uber/SnapChat/etc. I'm not sure how I'd alter your methodology to account for that, but it just seemed like a comparison that will always make the tech market sector look good, because the best companies are overweight in the index.

edward lamper said...

A query when we automatically Adjust Beta upward as the investment is concentrated to 1 asset and not many to account for only firm specific risk don't we account for Illiquidity already by pushing Risk Factor upwards and not factor in Illiquidity Discount later. Lot like the Brand Name High Operating Margins for Coco Cola.

edward lamper said...

Other query why should the public companies pass the value which they get from lower WACC due to lower Beta to the seller of small private company even if there are more than 1 public companies. Specially when the private company had no hand in providing that added value due to Financial and Operational structure of the public company interested in buying it.