The Shape of Bubbles – The Madding Crowd

I posted earlier in the week about my guesses as to how things shake up if (when) the current tech cycle bursts (like a Bubble). And I realized I had forgotten a key component of how things will fall apart once the business cycle turns. I remember this theme from the 1990’s Bubble, and it seemed so obvious in hindsight, but no one seemed to write it until after things had gotten bad. So I want to stake some ground here in the fortune telling business.

One of the key problems with inflationary periods of business cycles is that there is a lot of easy money floating around. This lets entrepreneurs take risks, but it also greatly lowers the barriers to competition. This results in too many companies chasing the same market.

Currently, many start-ups are in ‘investment mode’. They are burning cash today, in order to build a better business tomorrow. Some are trying to build up market share, or seed a market to reach critical mass, there are lots of examples. This is a sound strategy, but the problem is that it is a sound strategy for lots of competitors. When capital is cheap, as it is today, it can be very hard to determine if a business is burning cash because it is ‘investing’ so heavily or if it is stuck in an market with unsustainable economics.

Let me give an example to make this tangible – on-demand meal service. We use Munchery at home, and have really liked their model of delivering meals to our house in under an hour. They are reportedly raising $85 million. The problem is that almost every time I open Twitter I see a coupon for a competitive service. The same is true of  laundry pick-up services, toy-delivery services and many others. There is just not enough room for all of these in any city. We saw the same phenomenon in the 1990’s with a similar sector – urban delivery services, and there are plenty of other crowded fields.

Just to be clear, I do not mean to criticize the management team of Munchery or any of the others. Now is the time to spend money to build market share. There will be a shakeout someday and only one of them will survive. My point here is just that it can be very hard to remember this when the times are good. It is very hard to determine when things will turn bad and just who will survive. There is some strange version of the Prisoner’s Dilemma at work here – everyone is acting rationally but almost everyone will lose in the end.

For those of you who follow the public markets, you can already start to see this analysis working its way through the digital advertising stocks. Many investors want to short Internet stocks. They argue that the combined market cap of Facebook, Google and Amazon already exceeds the total market for advertising, The bears argue that advertising is ultimately a zero-sum game, and you cannot have all these stocks maintain their valuation indefinitely. The bitter irony is that none of the shorts can agree on who is overvalued. Facebook? Google? Apple? Amazon? Some company we have never heard of that is going to become tomorrow’s Unicorn? (While we are on the subject of easy capital leading to too many market entrants, has anyone noticed how many hedge funds there are out there…)

In the end, when the money spigot gets shut off, there will be a fair amount of suffering to go around. One of the biggest multipliers of that suffering will be the extent to which certain markets are over-saturated. Many of the companies in those spaces will have to shut down, and their employees will not be able to bolt to the competition because the competition has shut down too. Support staff will get hit as entire job categories disappear, and we will find odd pockets of pain distributed in unexpected places.

Today all of these look like healthy companies, but someday we will stop measuring the success of a company by ‘growth’ and then have to redefine what a healthy company looks like.

This does not have to be a tragedy or a disaster, just a reversion to the mean of business cycles.

What would cause the Tech Industry to slow down?

In my previous post, I skipped a step and made some guesses about how a slowdown in venture funding could ripple through the industry. Is such a slowdown likely? I would argue that it is inevitable, all industries ebb and flow. These tides of fortune are driven by capital flows more than anything else. And right now, capital is very cheap for a large number of Internet, software and other tech companies.

Capital is cheap because of global macro-economic conditions. There is a ‘savings glut’ that is so pervasive it has its own Wikipedia entry (and here is a good page of Google search results on the issue). Lots of people (or countries) have big piles of cash and are having a hard time finding returns on that money. When stock funds generate 3% returns, investors look to riskier assets – art, luxury real estate, private equity, venture capital, etc.

Someday, this will change. I have no idea when, but I think it is reasonable to suspect that whenever interest rates start to go up in the US, it will spark follow-on changes throughout the economy. It will not happen overnight, but eventually interest rates will get high enough on bonds and bank deposits, that investors will start to park money there rather than risk losses in more exotic assets.

I used to hear all kinds of arguments about how this will not affect the Valley and the tech industry – lean start-ups, capital efficiency, this time is different, etc. But lately, I think most people would agree that current conditions can not continue forever.

Instead, I think it is more productive to think of what leading indicators we can expect. I would not argue that anyone should pull back on venture investing now. This is still a great time for many companies to grow. There are a lot of very powerful, productive forces getting unleashed. Invest in them, grow them. But it would also be nice to have a bit of a heads up that the good times are going to take a pause. Some of these could include:

  • The closing of the IPO window. It is wide open now, but that will not last forever. This may come after a sharp ‘correction’ of the broad stock market. These things are not always directly related. You can have IPOs in bad markets, and no IPOs in good markets, but there is some correlation.
  • More than one venture shop failing to raise as much money as expected.
  • Craziness in the bond market. This is actually the one I would track most closely. The other two are pretty obvious, but no one in tech watches bonds. And this is probably the best fortune teller out there. A falling bond market, means interest rate expectations are going up everywhere. And that will make venture investing relatively less attractive.

The key to all of this is to not expect dramatic revelations. The drama only comes once things are obvious. In the past, I have tried and failed to predict when a slowdown will occur. I really have no idea when it will happen. But I think it is worth preparing for such a day.

This is not a post about Bubbles

This post is not about Bubbles, or the popping of Bubbles. If that were the subject, I would have to spend the first few hundred words arguing why I think we are in a Bubble, and let’s face it, that word has been used to death.

So this is not about Bubbles. This post is about business cycles.

The ‘tech’ industry (whatever that means) is mature enough so that it is past time we think of it in terms like that. The whole industry does not have to veer from boom to bust to boom again. Instead, there are expansionary periods and contractionary periods. The difference being the magnitude of the swings relative to the size of the whole industry.

I think its safe to say the industry has had a pretty good run for the past few years. And I believe that things will slow down a bit in the next year or two. This does not need to lead to doom and gloom, we are not talking about bubbles. But I do think it will mean slowing growth rates, a larger number of companies shutting down and a general wave of retrenchment.

So rather than get into an endless “Are we in a Bubble?” debate, let’s assume that sooner or later things slow down. What is that going to look like? I will do a follow-on post about what I think will cause the slowdown, but let me start with the assumption that a key factor will be a slowdown in venture funding, and that there will be concurrent changes to the broader global economy.

Before I get into who suffers most in such a downturn, I want to first look at some general patterns in such situations:

  • Downturns have cumulative effects. One company shuts down and its employees are out of work, so they cut back spending on housing and lattes. The company’s suppliers lose a revenue source and react accordingly, possibly shutting down themselves. This adds up, leading to…
  • Indirect outcomes are powerful and hard to predict. Management teams (and humans in general) have a hard to time making non-linear predictions. A company’s sales forecast is easy to predict, until it isn’t. Sales pipelines do not work well when customers start to disappear. Your customers’ customers are hard to track. This problem is made worse by….
  • The speed at which companies today operate cuts both ways. Companies have become accustomed to things changing quickly. In an upward cycle, this is a management headache and an opportunity. In a downturn, it means the number of ugly surprises multiply.
  • Pay Attention to Marginal Costs and Revenues. We have become accustomed to infinitely scalable Internet business models. The truth is that during inflationary periods assets get priced by the case at the margin, meaning the next one sold. This effect has been killing commodities like iron and coal. Small decreases in demand are having outsized impacts on their prices. Finding the equivalent in tech is not easy, but I suspect there will be many things (houses, companies, messaging companies lacking revenues…) where pricing is driven by peak conditions, and thus deflate faster than expected
  • A falling tide reveals who is naked. Put another way, hard times reveal true management ability. Many seemingly well-funded companies will collapse because their management teams are either spending much more than was known or they cannot make the transition to new operating conditions. Take a look around your company now and ask yourself how much of the spending is rational and how much is the result of having too much money in the bank?

With that all said, where do I think we will see trouble?

First, there will be some kind of shake-up among mid-stage start-ups. If the venture tap closes, there will be a wave of closures and fire sales among companies that were furthest from their last funding round.

Second, there will be a wide range of outcomes depending on the sector. Companies that are building enterprise sales platforms will struggle as they are likely further away from break even. Companies with no revenue model, hoping for a sale, will find their backs against wall very quickly. SaaS models and many Internet businesses will likely fare well, if they have reasonable burn rates. Here, I am thinking of companies like Uber and Airbnb. These businesses are highly capital efficient, and in theory can ‘flip a switch’ from growth investing to cashflow conservation. (I could write a whole other post on how this is not actually such an easy switch.)

Those are the direct impacts, but I think the bigger problems will come from indirect effects. Predicting the losers in a bad business cycle is always full of surprises, and part of the motivation in writing this note is to help me think through the less obvious victims.

So the third bucket of victims are companies that are recipients of start-up spending. If I were a venture investor, I would take a hard look at where my portfolio companies are spending their money, and then avoid investing in those areas. I think there are a number of sectors that are going to struggle. Even the best-funded companies can falter if their customer base evaporates. Here, I am thinking of the (far too) many ad tech start-ups and ad networks; developer tools and other developer focused companies. In addition, I think there are many service-oriented start-ups (insurance, catering, benefits, accounting, etc.) that begin life targeting other start-ups. While Zenefits is sufficiently funded for the next decade, anyone looking to follow in their footsteps may find a much smaller universe of early customers.

Fourth, I think that many companies from the ‘previous generation’ of start-ups will fall on hard times. Last week Zulily reported some bad numbers and saw its stock take a real beating. I do not see that as a harbringer of an e-commerce meltdown, but I do think it shows how quickly once-hot companies get surpassed. That process gets accelerated when the whole industry turns down.

Fifth, something bad will happen to the angel investing process. I do not mean that angel investing is bad, but it is one area that has seen some abuses, having grown too quickly. To be clear, I am really thinking of the large and growing number of angel investment groups comprised largely of newcomers to venture investing. I am sure Angel List will be fine, but there are now many other ‘amateur’ groups out there. My sense is that these groups will take a big hit, having over-funded too many companies with limited potential. I believe that it is too easy to get seed funding now, and this will inevitably lead to a shake-up.

Sixth, it is unclear if there will be a systematic impact on the broader economy. This is what made the 1990’s Bubble so pernicious, it covered so many sectors (telecom, media, Internet) that its bursting brought down the broader economy. Unless things get truly out of hand, I do not see that happening this time. However, it seems likely that any shakeout in the tech industry will also likely take place at the same time as shakeouts in other parts of the economy. I think risky assets everywhere will slow at the same time venture investing slows. These things will be concurrent but not connected, and the distinction may get lost.

Finally, I come to the inescapable conclusion that the biggest victim of a downturn in tech spending will be the San Francisco real estate market. As a homeowner here, the smart move would probably be to sell my house. That is not going to happen, but when the ‘tech industry’ does enter its contractionary phase, the bubble of San Francisco real estate will pop.

Jana Vs. Qualcomm – The Only Thing to Fear is Fear Itself

I had an odd conversation recently. Speaking to a friend at Qualcomm, he said that the management team there was really worried about Jana Partners’ activist attack on the company.

That did not make sense to me. I just could not see them getting too worried by something like this. But then I heard identical comments from several other people. And so now I am starting to believe that there is actually a bit a trepidation on Morehouse Drive.

I say this is strange because Qualcomm has become such a massive company, so powerful in its industry, that it should be able to fend off any activist. After all, activist strategies mostly boil down to massive head games played out in the press and in the markets. If you are nervous when you start the fight, then you are already psychologically defeated.

As I mentioned in my last post, I have some familiarity with shareholder activism from both the investor side and from the company side. I am no expert, but I have a decent sense of the strategies used to mount an activist attack and to defend against one. (I assign no moral judgment to either side.)

So it strikes me as odd that Qualcomm should be worried. They have so many factors working in their favors and levers to pull, that fending off an activist should not prove too difficult. Or at least, it should not keep anyone up at night. Moreover, Qualcomm is known as a company with a very high level of self-confidence (usually described by its competitors with a less flattering word). That trait seems to have deserted them as they go up against Jana.

Some of this concern may just come from the surprise factor. Qualcomm should have seen this coming, but they are not the first company to underestimate the scope of activists.

However, I think that some of their concern also stems from the fact that Qualcomm is starting to show some vulnerabilities. As I mentioned in my previous post, the company is now mature enough and large enough that built-in flaws have become apparent. All companies have these, Qualcomm is big enough that they are more readily visible to outside observers.

For instance, they have a problem with their margin structure. They are not earning as much as they should. In fact, part of Jana’s ‘advice’ is that the company undergo a big cost reduction. Finding those costs is not easy, but if you ask anyone who works there (and I have asked many), it is clear that there is ample cost to cut. And these costs can be cut without jeopardizing the overall business.

I will not get into all the details here (happy to discuss offline), but there are pretty clear ways to reduce costs at the company. Moreover, I would argue it is in the company’s best interests to do this. They would benefit from a bit more focus, discipline and energy.

And this leads back to why I do not understand management’s fear about dealing with an activist. Ultimately, everyone has the same goal – get the stock price up. Successful activist defenses basically boil down to nullifying the activists’ arguments by co-opting their strategies. In Qualcomm’s case, they do not need to adopt all of them. Certainly splitting up QCT and QTL is a mistake. But increasing their buyback and raising the dividend, those are not such bad ideas. The company’s long-term health would also be improved by taking the cost reduction steps necessary to make those financial strategies possible.

If I were advising either side of this battle, it would be to focus on cost and incentives throughout the organization.

Activists Target Qualcomm – And Get it Half Right

Last week, activist fund Jana Partners published a letter calling on Qualcomm to take several steps to improve shareholder value. (I know I am late in writing about this, but I literally have a note from my doctor with a pretty good excuse for tardiness.)

I do not want to wade into the whole subject of activist investors. I have lived on the front lines of both the ‘pro’ and ‘anti’ sides of shareholder activism, and it is far too nuanced to get into here.

So setting aside the big picture topic, I want to explore the specifics of Jana’s letter to Qualcomm. (I could not find an original source for the letter, the closest I have is this link from, someone let me know if I there is more that I am missing.)

Jana makes the argument that Qualcomm is undervalued relative to its importance in the mobile and computing industries. 100% agreement here, I have said repeatedly on this site, that no one seems to fully grasp how influential Qualcomm is over the mobile industry.

Jana then goes on to make six suggestions as to how Qualcomm can unlock that value. And this is where I start to see trouble. The big headline request Jana makes is for Qualcomm to spin off or sell its chip business (QCT), fully exposing the benefits of their IP Licensing business (QTL). This is a textbook ‘shareholder value’ move. And while it is often a very smart idea, I think Jana could not be more wrong in suggesting it here. These two businesses are joined at the hip, that is the secret to Qualcomm’s success.

Qualcomm’s model for the past decade (and probably the coming decade) is simple. The company works closely with the telecom operators, spending a lot on R&D to develop new features (and saving the operators the cost of having to do this themselves). The operators then vote these features into the mobile standards (aka 4G, LTE, 3GPP, etc.). Once the standard is set, the mobile handset vendors have to find chips that meet the standard, and who is the first company to offer those chips? Correct, the company that helped design the standard in the first place.  All the R&D that goes into supporting the new standard becomes intellectual property (IP) in the form of patents, and those patents are the basis of the company’s QTL business. The chips and the licenses are thoroughly intertwined.

Of course, there is a lot more complexity to this. Also, I have had to choose my words carefully here because this is one of those subjects, where if you stare at it too long, and squint a little you can start to see regulatory implications. This is going to make it hard for the company to defend itself. But it should be clear that the two businesses benefit from being together. It is also unclear what would happen to these two separate entities. Jana makes the suggestion that maybe Intel would buy a spun-off QCT. And I have to admit, that I laughed out loud when I read that part.

That being said, I actually agree with most of the rest of what Jana is proposing. Qualcomm is starting to show its size. Big companies all expose problems that come with time and age, but each in their own separate way. Having spent a lot of time in San Diego lately, I have a pretty clear idea of what particular form of sclerosis afflicts Qualcomm. There is something not right about their cost structure. Some of it comes from making some big bets – pushing process nodes, the $1 billion a year they are spending on ARM servers, etc. But they also have some serious organizational structure issues. I will not go into details here, but is pretty clear that the company is not staffed efficiently, and that they could greatly enhance ROI and time-to-market through some of the changes Jana touches on. There is ample room for improvement.

Qualcomm is an important, powerful company. It is not going away, and it does not need to split itself up. And maybe that is the problem. It is hard for big companies to undertake structural improvements, and doubly hard when the company sits in such a strong position competitively. Qualcomm could probably coast for a decade on its current position, but maybe having an outside agitator can spur the company to make real change.

Let me close by saying that I do not own any Qualcomm shares. I almost bought some after Mobile World Congress because the sell-side has gotten almost 100% negative on the name. I have hesitated because I am aware of all the cost and organizational problems I mention above. It just looks so messy when you get up close. Nonetheless, I think Jana is on to something here. While I disagree with some of their prescription, I agree pretty strongly with the overall diagnosis.

Retail Real Estate Stocks Scares Me

The real title of this post is “Is the market under-pricing technology obsolescence risk?”. Which is probably the most boring title I have ever written. So I went with the flashier title you see, because I think I am on to something worth pursuing here.

In my last post I laid out a theory that the Stock Market is mis-pricing the risk of a company missing a product cycle and becoming obsolete – hence technology obsolescence risk. I won’t get into all the mechanics, it gets drier than the headline, but I now have this theory that the stock market is penalizing certain companies with too high a risk of going obsolete. (And under penalizing those that are already over the shark.)

I am trying to figure out a way to prove that, but it requires a lot of data. (One person did offer me access to their FactSet feed and it is taking me a long time just to compile the right query.) I think that if we could assign a measure of tech risk to a stock, we could build a better index of technology stocks, that would outperform any index that included all those out-of-date companies.

Then I went for a drive in the desert.

In particular, I visiting family in the High Desert of Los Angeles. Unlike the rest of Southern California, the town of Lancaster knows that it is a desert – windows are tinted, stores are heavily air conditioned. And one day, when it was 90+ degrees outside, I ran some errands and visited a bunch of big box retailers – Toys ‘R Us, Best Buy, etc. And they were all empty. This was the week before Easter, which has apparently become a big toy- and gift-giving holiday. Despite being veritable oases of air conditioning, almost no one was shopping.

I had to think that as climate controlled as those stores are, an even better strategy for shopping in the desert is to make heavy use of Amazon. (Disclaimer: I own 100 AMZN shares.)

Now, I do not follow retail or real estate stocks much. And I am aware that investors in those sectors fully appreciate the risk that Amazon and Internet retail pose to the broader market. But I have to wonder, if even tech investors have a hard time pricing the risk of obsolescence, it must be much harder for investors who do not see that much in their area of expertise.

The thing that always surprises me in tech is that when a company falters, it fails very quickly. I mentioned Nortel last post, but think of Motorola, Nokia and Blackberry. All went from top of the heap to the bottom in under eighteen months, once things turned. Could we see something similar in retail? My guess is that retail investors are already familiar with this sort of problem – Radio Shack being just the most recent failure.

But what about real estate investors? I am sure real estate investors are aware of the Internet and the fact that it can hurt many potential tenants. But is that risk truly reflected in real estate stocks. I have to think it is not.

This train of thought was reinforced recently when I spoke to the channel manager for a large electronics OEM. Apparently, his team had recently conducted a large review of retail trends, forecasting out the landscape five years from now. His conclusions were sobering. His team is now taking big steps away from relying on distribution through physical retailers. He was incredibly downbeat about the future for electronics retailers. His view was that the Amazon and the other e-commerce players were lightyears ahead in their ability to use data to drive sales, and were only just scratching the surface of what they can achieve. This did not sound too promising for physical retailers.

Again, I think many investors are aware of this trend. But I think they are going to miss two things.

First, the transition away from physical retail is accelerating and will likely come sooner than many expect. In the technology industry we are accustomed to the notion that things move very quickly, and the risk of corporate death are high. This is not true in other industries, and it is easy to overestimate the amount of time remaining.

Second, this problem extends way past electronics. Home Décor, collectibles, food and most important – clothing. I spent some time in the Desert Mall as well, and by my rough math, 85% of the stores there were exposed to growing Internet distribution. People still like to try on their clothes before buying them, but that is very much a cultural preference, which is fading. For years everyone assumed that categories like luxury goods would be immune to Internet shopping, but Richemont’s acquisition of Yoox last week means that even that sector is in play.

What will happen to all the malls and big box retail shells ten years from now? I am sure that physical retail itself is not going to become obsolete. Teens everywhere will always want to go hang out at the mall. But I have to imagine that physical retail sales volumes are going to shrink to the point that many physical locations will have to close. And there is the very real risk that this happens faster than anyone is currently expecting.

Should private companies raise money from mutual funds?

There is a trend lately for private companies, in later stages of their fundraising, to include mutual fund managers in their venture rounds. In theory,the idea has a lot to offer, but in practice this is one of those things that does not always work out as well as planned.

It seems like it should work. The big mutual funds, firms which typically invest in public securities, devote a small fund to late-stage public companies. It gives the mutual funds a chance to improve returns slightly. Venture returns seem to be on fire lately, and the big mutual funds already invest in a wide array of securities to generate better yields for their own investors (i.e. you, me and everyone else who has a 401(K) plan). It also seems like a good idea for private companies, a way to dip their toe into the water of eventually going public and dealing with public investors. As one CEO put it to me recently, it is a way to “inoculate” management teams to the vagaries of the public markets.

This idea is not new. I can remember two prior periods when public market investors tried something similar. The first was during the 1990’s Bubble, and then again in the mid 2000’s. The first time it was mutual funds, the second period saw a lot of hedge funds make a similar attempt. There were probably periods that predate those. And if you look at the timing, the motivation of the public investors is clear. They are periods when the big funds had ample cash to invest, and periods when the venture market seemed to be on fire. Neither of those attempts ended particularly well.

The problem is that for the big, public funds venture investing is a very small piece of their overall investment portfolio. As such, it never seems to get the full backing of the institutions behind them. This creates a seemingly inevitable mismatch in timing between when the funds want to exit and when the companies want to provide them that exit and go public.

Typically, the big funds hire very good, very smart investors (not always, but often) to run their venture books. These managers make smart bets, but eventually the timing issue comes up. Funds that buy public securities need to have room to maneuver, they are accustomed to the fact that they can exit a stock or bund investment in the public markets at any time of their choosing. For the big mutual funds, their ability to exit a stock becomes a cap on how big a position they are willing to take in a stock. These investors look at a public stock and its daily trading volume. Usually, they do not want to buy a position that is more than seven times daily trading volume, and even that level is pushing it. So if a stock trades 100,000 shares a day, the fund manager who takes more than a 700,000 share position in that stock is going out a career limb.

By contrast, a private company CEO is accustomed to venture investors. They expect to eventually exit, but are willing to wait years, not days to do so. For a company in late-stages before an IPO, the last card the CEO has to play is the timing of that IPO. The decision to list will hinge on many factors beyond just the demands of one or two investors. By the time a private company gets to the size where these things happen, they typically have close to two dozen investors. All of them want an exit, and if only one or two sees urgency in going public, the company can hold off for a period.

What seems to happen next is that there is some shift in tone in the public markets. The markets crash or interest rates change direction. This causes the investment strategy committees at the public funds to rethink he way in which they allocate funds among the various classes of securities. And suddenly, the team running the private funds find that their investment horizons are shortened. Suddenly, they are under immense pressure to get liquidity in their investments.

At this point, something has to give. Either the fund manager brings so much pressure to bear that the company is forced to go public early, which makes the company unhappy. Or, the company delays its IPO (I mean, look at those markets…), and the fund manager suffers ugly returns, which makes the management company unhappy.

That, at least, is the way this has played out in the past. Maybe this time is different. The biggest mutual funds are now very big and have all sorts of less liquid investment on their books. And if I remember all of this, the people running the fund companies can certainly learn from history.

Still, I would caution private companies from taking funds from public investors too early. I caution all management teams about taking any outside money. Taking an investment means you are going into business with those people. Most companies are now well-practiced in picking their VCs carefully, selecting those that offer advice, access or some other strategic benefit, over those merely offering cash. The same holds true for taking money from mutual funds.

And as for the ‘inoculation’ factor, it only goes so far. There is very little training to prepare a CEO or CFO for dealing with the public markets, other than ‘on-the-job’ training. Having a few, venture-like public investors is still a far cry from dealing with a sea of hedge funds. The public funds tend to send people with venture or private-company backgrounds to run their venture investing, and as such, they are a very different personality type than the majority of public company investors.

I do not think there is a hard and fast rule here. Sometimes these investments go exactly as planned, but just as often they do not.


Get every new post delivered to your Inbox.

Join 748 other followers