What this covers
This April 2020 conversation features Gavin Baker on how an investor should operate inside the COVID-19 selloff. He treats the downturn as something without a model. It is not a financial crisis in which credit problems spread to the real economy. The economy simply stopped, producing the sharpest GDP decline on record, so outcomes are inherently unpredictable. His answer is discipline about the present: drop anchors from past views and recent mistakes, keep only broad principles, and borrow Bruce Kovner's image of a boulder at a mountain top whose every path down must be imagined, a way of planning for path dependence.
The middle of the conversation draws on history and market mechanics. Baker cites Martin Fridson's study showing that the ten most exceptional stock market years all followed terrible periods, with 1919, after the Spanish flu, the exception. He says Market Internals, meaning aggressive names beginning to beat defensive ones, tend to turn before the bottom, as they did before 2009. He also argues, contentiously, that covenants matter less than cash because lenders avoid forcing bankruptcies, except for revolvers. He adds that PayPal's fate hinged on raising money just before the 2000-01 collapse.
The last stretch applies this to specific businesses. Recessions, he says, accelerate change, here in video, social networking, and e-commerce. He doubts Facebook's Oculus bet on the Metaverse, since video games such as Roblox may become the platforms. He defends WeWork's flexible-space concept for corporate branch offices. He sees iBuyers as market-makers whose real risk is inventory they cannot reprice.
Baker argues the COVID-19 bear market is an unprecedented opportunity for alpha-focused investors, who should avoid anchoring, rely on cash over covenants, and position for structural shifts the recession is accelerating, while expecting heavy government support to limit the downside.
- Peak volatility, peak valuation spreads, and high correlations together create the richest alpha opportunity in about 100 years
- Liquidity and cash matter more than covenants, and days of solvency with zero revenue is the key test
- Recessions accelerate change, so companies whose competitors fail or whose behavior shifts (Zoom, omni-channel retail, video games) stand to gain
Extreme dispersion and small UI/UX differences create the biggest alpha opportunity.
- Apple with FaceTime, Google with Hangouts, and Facebook with its video chat missed the biggest change from this recession because their video products were not as good as Zoom on UI and UX.
“Instead, I think they've already missed one of the biggest changes to come out of this recession.”
- Peak volatility, peak valuation spreads within sectors, and very high correlations combined make this the biggest alpha opportunity for long/short and long-only managers, statistically as rich as any in the last 100 years.
“this is statistically as rich of an opportunity set as we have seen in the last 100 years”
- Absolute return investors care about absolute valuations rather than valuation spreads, so the extreme dispersion in this market matters far less to them than to alpha-seeking long-only or long/short managers.
“If you're an absolute return investor, it's a little bit of a different environment because you don't care as much about valuation spreads. You care about absolute valuations.”
- E-commerce will structurally increase its share of retail sales, but it may not take share faster than the consistent five-to-six-year trend, and it will likely be more competitive in 2021 than in 2019.
“I don't know though that after the structural shift up, it will continue to take share at anything different than the rate it's been taking share for over the last five to six years”
- Zoom is the most intellectually interesting stock because it has become a verb, and companies that become verbs generally do well for long periods, though nothing is forever, as Xerox shows.
“companies that become verbs generally do well for a long period of time. Google, Xerox. Xerox obviously did well for decades and then eventually imploded.”
- Tiny UI and UX differences matter immensely, since a 50 basis point user-experience advantage can underpin hundreds of billions in market value, and Google found that shaving two milliseconds of search latency measurably affected search volume.
“very small differences in UI and UX matter immensely... 50 bps is huge. If you're 50 bps better, that can be the foundation of hundreds of billions of dollars in market value.”
Unprecedented downturns require holding many scenarios and wide perspectives rather than a single forecast.
- This is not a financial crisis where credit problems transmit to the real economy; the economy simply stopped, producing the largest and sharpest GDP decline in history with no model or precedent, making outcomes inherently unpredictable.
“We have the biggest, sharpest GDP decline in history... This is not a financial crisis where credit issues are transmitted to the real economy. The economy just stopped.”
- Bruce Kovner imagined himself as a boulder at a mountain top and visualized every path it could take down, because path dependence and sensitivity to initial conditions mean planning for many scenarios keeps you ready for whichever one unfolds.
“he tries to imagine all the different scenarios of the Boulder traveling a different path down the mountain because there is a lot of path dependence and sensitivity to initial conditions.”
- In this fog-of-war environment, the only reliable information is trailing financial numbers, which are actual results, while forward estimates mean nothing, and every company will miss and guide down.
“the only information you can rely upon is information that is wildly stale. And that is trailing numbers... no forward number means anything. Every company is going to miss. They're all going to guide down.”
- The debate over recovery shapes (L, U, V, R, W) is alphabet soup, and investors should hold a wide range of scenarios rather than commit to one shape.
“is it an L, or a U, V, R, W shape recovery. You want to have a wide range of scenarios in mind.”
- Risk management requires being like Augustus, who learned from his uncle Julius Caesar's fate and was cold-blooded and dispassionate, rather than like Julius Caesar, who was brilliant but a risk-taker who died before 50 by walking unguarded into the Senate.
“When you come to managing risk, you want to be Augustus, not Julius Caesar. You want to be super cold blooded, super ruthless, super dispassionate.”
- For a growth investor the biggest single risk is a regime shift in valuation, where companies move from being valued on EV-to-sales to being valued on GAAP price-to-earnings, and this is hard to capture quantitatively.
“the biggest single risk to me is a regime shift in valuation... companies go from being valued on EV-to-sales, to being valued on GAAP price-to-earnings.”
- Almost no one could have imagined a virus shutting down the world, even though pandemics had been written about, which Baker attributes to human behavioral biases.
“nobody could have imagined what we're living through two weeks ago... I think it's down to so many human behavioral biases”
- Society and leaders must behave very differently in long winters than in summers, and we are now in winter, so investors and CEOs need to make harder decisions and think and act differently.
“I just think we're no longer in the summer, we're in the winter now. As an investor, that means you need to think and act differently.”
- Risk and valuation are like the parable of the blind men and the elephant, so investors should examine both from as many perspectives as possible and keep a wide aperture on the range of outcomes.
“risk and valuation, they're both the parable of the blind men and the elephant. You want to look at risk and valuation from as many perspectives as possible.”
- Investors are trained to almost never do what is obvious, because doing obvious things rarely generates alpha, which makes the timing of obvious-trade payoffs in bear markets particularly unusual and hard to act on.
“As an investor, you're trained to almost never do what's obvious. It is very rare that doing really obvious things generates alpha.”
Recessions accelerate existing digital trends like video, social networking, and e-commerce.
- Recessions are always accelerants of change: they alter both the speed and the direction of change, so the coronavirus recession is accelerating trends such as video, social networking, and e-commerce.
“Recessions are always an accelerant of change. They always change the vector of change - both the speed and direction.”
- TikTok is beginning to affect incumbent social networks for the first time, and every generation may end up with its own social network, as Snapchat's age barrier suggested.
“TikTok has begun to have an impact on incumbent social networks for the first time ever... I'm beginning to wonder if every generation is going to have its own social network.”
- Omni-channel retailers such as Walmart, Target, Costco, and Kroger are building years of e-commerce capability in months, and the internal debate about e-commerce investment has been settled in favor of going all in, which may hurt pure-play e-commerce companies.
“they are building years of kind-of eCommerce, experience, muscle know-how in months... That debate has been settled, I would imagine.”
- Internet traffic data shows video traffic up 100 percent per Verizon, and Telecom Italia reported video game traffic up 75 percent while social media traffic was flat, suggesting people are connecting inside video games during lockdown.
“Video traffic is up 100% as people are working from home and kids are spending more time at home... Telecom Italia saw video game traffic up 75% and social media traffic up 0%.”
Winning tactics shift by market phase, favoring obvious trades early and non-obvious risks later.
- Investors in a bear market must play in the present, not anchor on what they thought they knew or on their own recent mistakes, keeping only broad principles, because mistakes are inevitable and must not impair future thinking.
“you cannot anchor... you have to play in the present. You cannot let the mistake that you've just made impair your thinking.”
- Bear markets work on a FIFO principle: the first companies sold are the first to be bought back, and defensive names with great balance sheets get shot last, so how you invest changes with the phase of the market.
“Bear markets, they work on a FIFO principle, not LIFO principle. The first companies to, kind of, be sold by the market are the first ones to be bought.”
- An active investor in a bear market can either do nothing and trust strong balance sheets and competitive advantages, or trade actively, which Baker compares to a knife fight while blindfolded and covered in grease.
“Either you can do nothing, believe in your companies... Or, you can trade really actively. And if you're going to trade actively, I wrote on Medium that it's like being in a knife fight when you're blindfolded and covered in grease.”
- The obvious bear market trades have been taken, so generating alpha in the late stage and recovery requires taking risks that are not obvious, since the first stage paid for obvious actions.
“If you want to generate alpha in the later stages of a bear market and particularly in a recovery, you now need to begin taking some risks that are not obvious.”
- At the front end of every bear market, doing the most obvious things works, as shown by investors who sold airlines, travel, and cruise names and bought Target, Walmart, Costco, Kroger, Amazon, and Netflix in the first three to four weeks, even though this was hard to act on.
“at the beginning of almost every bear market, doing the most obvious things works. So, who crushed it for the first three to four weeks of this bear market?”
- The best stocks coming out of June 2009 included Netflix, Booking.com, Ulta Beauty, and Domino's, with Domino's a heavily levered post-LBO name that had been hit hard.
“go back to what were the best stocks coming out of June 2009? Netflix, Booking.com, Ulta Beauty, Domino's pizza. Domino's was a name that was incredibly levered.”
- Market internals show aggressive names beginning to outperform defensive names, which generally happens before the market bottoms, so holding aggressive names gave Baker positive alpha well before the 2009 bottom.
“what that basically boils down to is the aggressive names start to outperform the defensive names. And that happens, generally, before the market bottoms.”
Optimism and buying after terrible periods have historically paid, and recession timing differs from restart timing.
- Companies that recently completed financings or large divisional sales had financial statements that did not reflect the new cash, so the market traded them as if they lacked billions in cash, an inefficiency Baker actively exploited.
“companies that had done a financing in January or February, or had sold a big division, their financials were obviously incorrect... it was trading... as if it didn't have that $2 billion in cash.”
- Optimism has consistently paid over the last two to three hundred years of human history despite wars, Spanish flu, depressions, and nuclear threats, so it is good to bet on humanity even though optimism never sounds smart.
“the history of the last two to three hundred years of the human race tells you that optimism has paid consistently. It is good to bet on humanity.”
- Martin Fridson's book on the 10 most exceptional stock market years of the 20th century shows that they all came after truly terrible periods, though 1919 was not exceptional after the 1918 Spanish flu, while many exceptional years followed periods at least as hard as 2020.
“they all come after a truly terrible period... a lot of the exceptional years came after years that were at least as hard as what we're going to go through in 2020.”
- PayPal would not be the company it is today had it not raised money just before the market imploded in 2000-01, so the timing of fundraising, especially in venture, can determine the fate of companies and industries.
“PayPal would not be the company it is today if they had not raised money right before the market imploded in 2000 or 2001. So, the timing of fundraising is going to kind of rule the fate of many industries”
- The world beginning to restart does not mean the recession is over; it means Americans slowly return to normal daily patterns of eating out and going to stores, so the restart and the recession end are separate events.
“by the way, the world beginning to restart does not mean that the recession is over.”
Aggressive fiscal and monetary policy will stop the recession, though small-business aid may arrive too slowly.
- Stimulus will be increased until the recession stops, and the political willingness on both sides of the aisle to do whatever it takes is very high, so the government can and will act to stop this downturn.
“the stimulus will be increased until the recession stops... the government can and will do anything to stop this”
- The Small Business Administration processed only about 58,000 loans last year and employs only a few thousand people, so it is doubtful the small business loan program can flow quickly enough to matter.
“can they really flow through the small business administration, which I think processed only 58,000 loans last year and only employs a few 1,000 people?”
- Sending checks directly to people is a radical policy that economists have long advocated, but every day and week matters, so the checks must arrive quickly to be effective.
“he has embraced the idea of sending checks to people. This is something that economists always advocate... But every day counts, every week counts. Those checks need to hit quickly.”
- Policy response is the most aggressive in history, with more fiscal and monetary stimulus than any prior recession, the most aggressive fiscal stimulus in at least 50 years, and it is moving faster and more coherently than in 2008.
“this is certainly the most aggressive fiscal stimulus in the last 50 years. And more importantly, it is happening fast”
Recession winners are omni-channel retailers and quick-service chains as small rivals fail.
- A dentist was advised by an industry webinar to call software vendors and threaten to switch unless they cut payments, and she cut her software bill in half in one hour with zero pushback, which Baker reads as an indicator but cautions is only anecdote.
“In one hour, she cuts her software bill in half. One hour. She said there was zero pushback.”
- The companies that benefit most from this recession are those whose competitors go bankrupt, and Baker is focused on omni-channel retailers and quick service restaurants because many small retail and restaurant owner-operators are likely to fail.
“the companies that are going to benefit the most from this are the companies where their competitors go bankrupt... a lot of small businesses in retail and restaurants are going to go bankrupt.”
- When the coronavirus quarantine ends, people will first return to quick service restaurants with drive-throughs, and those chains will face less competition because many small family-owned competitors will have closed.
“I think the first thing that people are going to do, when this current coronavirus quarantine ends, is go to quick service restaurants that have a drive-through.”
- Facebook and Google ad auctions favor strong incumbents in strong economies but disadvantage them in weak economies, while advantaging smaller e-commerce operations inside large physical retailers, creating crosscurrents in e-commerce.
“the way that Facebook and Google auctions work, it really advantages strong incumbent players in a strong economy but it actually disadvantages them in a weaker economy”
Cash runway and liquidity matter more than covenants or leverage metrics in a crisis.
- Covenants matter less than cash on hand because debt holders do not want to force companies into bankruptcy, especially in a crisis, so liquidity is what counts; the exception is revolvers, where covenants really matter.
“covenants don't really matter. Cash dominates covenants. The debt holders actually don't want to put companies into bankruptcy... The covenants on revolvers really, really matter.”
- Heavy trading volume does not mean liquidity: futures bid-ask spreads widened to a single bid and a single ask at wildly different levels, so the market has minimal liquidity despite high volume, which makes limit orders critical.
“volume is not liquidity... there was one future bid and one future ask... although there's a lot of volume, there is minimal liquidity. So it is critical in this environment to use limit orders.”
- The new liquidity test that investment firms are running everywhere is days of solvency with zero revenue, rather than the Altman Z-score, debt-to-EBITDA, or interest coverage, because every company without revenue eventually runs out of cash.
“It is days of solvency with zero revenue. That is the equation that is being run everywhere. I don't know that this equation has actually ever been run before.”
Software revenue faces customer budget cuts, and work-from-home gains will reverse.
- Vista Equity CEO Robert Smith's line that software is better than first lien debt may not hold in this recession, because if first lien debt holders are not being paid, software bills may not be paid either.
“Software is better than first lien debt. And I'm just not sure that that is going to be proven true.”
- Software was 7 to 10 percent of IT spend in 2008-09 but is roughly 30 percent today, and including cloud infrastructure software plus cloud is almost 50 percent, so software now faces budget cuts that 2008-09 software did not.
“in '08/'09 software was 7% to 10% of IT spent. And today it's roughly 30%... software plus cloud, it's almost 50% of IT spending.”
- Software enabling work from home is struggling to keep up with demand and benefiting immensely now, though this advantage will reverse when the world returns to work.
“Any software company that enables work from home is struggling to keep up with demand. Now, obviously when the world goes back to work, they're going to be in a very different situation.”
- There is no such thing as truly recurring revenue, only revenue that is more or less recurring, and software revenue is exposed to customer budget cuts.
“there is no such thing as truly recurring revenue. Some revenue is just more recurring than others.”
- Software companies will likely not force customers into bankruptcy over unpaid bills, because they fear government intervention such as a holiday on software expenses, so they will extend terms instead.
“When push comes to shove, the software companies are going to blink... the government can and will do anything. They can just say, 'We're going to have a holiday on software expenses.'”
The Metaverse will be built on video games and platforms with identity and payments.
- The strategic control point of the Metaverse is the abstraction layer that becomes a platform, and Baker doubts Facebook's bet that owning the VR company Oculus will give it identity, avatars, and payments, since video games themselves may become platforms, as Roblox already has.
“I'm not sure that is true. And I think that there is a very real world in which some of these video games emerge as platforms unto themselves. So, this has already happened with Roblox.”
- Applying Occam's Razor, video games are likely to be foundational to the Metaverse and its social networking layers, since they already host concerts like Marshmello's 40 million-viewer Fortnite event and Star Wars premieres.
“I just think Occam's Razor is powerful... video games are going to be super important to it”
- Platforms with an identity and a payments layer, such as Xbox, PlayStation, and Steam, are well positioned to become Metaverse platforms, and Discord could go between work and play as a foundational platform.
“any video game platform that has an identity and a payments layer is fairly well positioned... I think Discord is something that could easily come out of this as a platform”
- Parents may come to see goal-oriented video games played with friends as a relatively healthy social activity compared with Instagram ego damage or passive TV watching, which could pull the Metaverse forward.
“parents realizing that relative to having your child's ego destroyed on Instagram... playing a goal-oriented video game with a group of their friends and socializing is actually a relatively healthy activity.”
- The Metaverse, a series of connected virtual worlds in which a majority of people will spend most of their waking hours within Baker's lifetime, is the culmination of the internet in the same way mass production culminated the industrial revolution.
“The Metaverse is simply a series of connected virtual worlds that I firmly believe a majority of people will spend a majority of their waking hours in, within my lifetime.”
The crisis will quickly settle contested tech business models like WeWork, iBuying, and live sports TV.
- The crisis will quickly resolve many hotly debated tech-enabled business model questions, including Space as a Service, iBuying, and fintech lending versus traditional banks, within the next few months.
“And they're all going to be resolved. We're going to know the answers to some of the most hotly debated topics within the world of technology investing over the last 10 years, and we're going to know those answers quickly.”
- The Ian Schrager-style model of packing more people into smaller offices with common spaces, which drove revenue per square foot in co-working, will have to change because containing the virus makes population density a priority.
“That model is going to have to really change in a world where containment of the virus is a priority.”
- WeWork's Space as a Service concept has been unfairly maligned, because corporate real estate heads found that flexible co-working for branch offices made sense, since branch footprints are rarely sized correctly and local knowledge is limited.
“WeWork has probably been unfairly maligned... Branch offices were always too small or too big.”
- Whether sports is the anchor of the linear TV bundle will be determined in the coming month or two, when live sports return or do not.
“Another thing much debated is sports really the anchor of the linear TV bundle. Well, we're actually going to find that out, this month and next.”
- iBuyers are essentially market-makers that earn a spread, so in financial stress they can widen spreads, but their real danger is holding large inventory they do not understand and cannot easily reprice in a recession.
“they're just making a spread, they're just a market-maker between buyers and sellers. And if that's the case, you just widen your spreads at a time of financial stress.”