• Alumni

Activity Feed

On November 20, 2016, Lee commented on How to Train Your Driverless-car :

I agree with the inevitable move towards self-driving cars and I can confirm that it’s now possible to drive a Model S from New York City to Boston without touching the steering wheel or pedals for over 90% of the distance. However, we are still quite far away from typing the destination address into the navigation system and falling asleep. The Google self-driving car is supposedly much closer, but still runs into problems such as staying “trapped” behind an active-duty garbage truck and requiring over an hour to drive what should have been a 10 minute local commute.

Suppose regulation does entertain allowing fully-autonomous vehicles on the current local roads. I’m interested in solutions to what could be a long phase where roughly half the cars are autonomous and the other half are still fully human driven. Does anything stop a wreckless driver from being even more wreckless, knowing that an autonomous vehicle will passively yield? Can pedestrians freely J-walk knowing the Tesla is certain to stop?

One solution I would propose is to equip all autonomous vehicles with not just an internal log, as you mention in the post, but a way to send the video camera footage to the proper authorities to report human traffic violations (the aggressive car or pedestrian should be ticketed, and there could be many self-driving car “witnesses” with camera footage to send). Of course, privacy hawks would be very unhappy if personal cars were suddenly turned into government surveillance tools. However, just as dashboard cameras on police cars improved police accountability, I would argue the safety benefits far outweigh the drawbacks.

As a heavy user of both SPG and Airbnb services, I believe the role of each company is very different. Airbnb likes to paint a romantic picture of saving households with financial troubles by helping to monetize the couch or spare room. However, we see more evidence that by dollar volume, the real superhosts on Airbnb are merely converting investors from a sublet motel to a short-term hotel model. http://avc.com/2014/12/what-just-happened/

We also see supporting evidence that Airbnb is moving away from the local immersion experience. They’ve recently created the “suitable for business travel” search features, and asked all of their hosts to update the amenities that reflect business travel. https://www.airbnb.com/business-travel

I agree, the industry will always give a nod to some local customs: asian hotels usually provide free breakfast, fancy western hotels offer turndown service, and SPG elites will be offered a local gift. However, the digital transformation and consolidation of hotel owners is doing more to standardize hospitality across diverse cultures into a more uniform experience.

We all hate paywalls and thankfully the current nature of the NY Times paywall is rather porous. http://avc.com/2011/08/on-porous-paywalls/

The big changes to digitial news consumption would occur when NY Times and other content providers all decide to finally close all the loopholes and put up paywalls (The Google and Facebook trick stop working, and all the major reputable papers require some payment). Would we see society move away from the $0 baseline that prevents so many of us from paying? Or would we reach a model where new entrants could come in at the free price and suddenly become a popular news brand, dislodging decades old publications such as NYT?

Either way, I believe journalism provides many positive externalities on society that benefit far more than the buyers of the daily paper. News outlets have been essential in uncovering cases of political and corporate misconduct. They have at times given a voice to the less powerful, leveling the playing field with free speech. I am all for finding a future business and operating model where deep, investigative journalism can still thrive.

My inner geek wants to believe drone delivery will be huge, and many others seem to agree given the number of drone startups coming out of the most recent Silicon Valley accelerators. https://techcrunch.com/2016/08/23/yc-demo-day/

I have heard a few arguments against the inevitable drone revolution. While I agree drones are a hugely useful technology, I wonder if image we have of a sky filled with little flying amazon boxes is a bit misguided in the short term (much like the Jetson’s depiction of a sky filled with moving sidewalks and personal flying cars).

For example, let’s examine very urban cities. The building architecture creates many unpredictable wind tunnels that seem to make drone navigation especially challenging and greatly increase the costs of accounting for flying condition variability (and related accidents). Also, for taking delivery, we currently require a secure place to store the package, whether it’s at a local aggregation center or with your building doorman. Would it make more sense to rethink how we handle last-mile package distribution and storage, as these problems still exist when we airdrop them at the require address?

On November 20, 2016, Lee commented on GameStop – A casualty of digitialization? :

Great analysis on the demise of the traditional GameStop business model. I would not be surprised if they shutter most retail stores in favor of the new digital models (assuming they survive the transition). One under-emphasized points transforming gaming is the concept of in-app purchases.

In-app purchases are gaming content add-ons for purchase to enhance the base game experience (the price of the base game sometimes being free). The new content can include extra levels, more features, or purely cosmetic alterations (a more interesting looking weapon or uniform in a multiplayer game, with no additional gameplay functionality). While the in-app purchase practice is still undergoing controversial changes, at the very least we know it moves many of the economics out of the hands of the video game distributor and back to a direct communication channel between the game publisher and game player.

On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

Regulation in the real estate industry has had a huge impact, both good and bad, in this country. We often take this for granted, but the Civil Rights Act and Fair Housing Acts have been essential to allowing equal opportunities in housing (short story, landlords and sellers previously practiced outright and flagrant discrimination; now it is at best subtle or circumstantial). Other laws exist to protect consumers from scams.

So while not all real estate regulation is bad, I agree there should be more requirements to give consumers access to crucial inventory and transaction data. There was a famous lawsuit in the early 2000s between software vendor MLX and the Real Estate Board of New York (every major brokerage firm and competing vendor in town was ultimately named as a defendant), disputing the anti-competitive practices of the industry group restricting which vendors were allowed to access, join, and participate in the intra-brokerage data sharing procedure. After years of bitter fighting, they finally reached a cash settlement and given access, but by then MLX had lost almost all of their clients and were soon forced to shutter operations. MLX was founded by Lala Wang (HBS ’78), who is a RentHop investor and adviser.


On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

You are correct that there is a huge distinction between a relocating client and a current resident of the city. Traditionally rental brokers love relocating clients. They appreciate the neighborhood expertise more, they often have a higher budget thanks (possibly with the employer covering broker costs), and most importantly, they actually have a set deadline for moving and finding a new home. Current residents almost always have a lower conversion rate because they can simply stay and renew whatever arrangements they already have.

The emphasis on relocating clients highlights one of the reasons cracking the chicken and egg problem, aka bootstrapping the networking effect, is so difficult. By definition, a relocating client only needs to use an online service once, and never again until the next city relocation. That means for a startup, product diffusion by word-of-mouth or brand equity is very slow, unlike say Airbnb and Uber, which are services that draw repeat customers multiple times a year. Real estate startups rely heavily on paid media, earned media, SEO, and partnerships to drive new traffic, and sadly many of those are zero-sum games. Only one website can be at the very top of the most relevant Google searches.

On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

I have always liked the idea of utilizing comparative advantage and price discrimination schemes to lower overall costs of broker services. To extend your idea further, why not hold a mini-auction among a marketplace of brokers the moment a buyer is interested in a property? For example, you found an apartment online and 6 different real estate agents claim to be experts in that neighborhood. Could you hold a reverse auction, where the agents bid on the commission they will charge you if you ultimately buy? That way the agent who lives across town might bid a normal commission but the agent who lives upstairs in the same building will only require one third of the commission?

So far a lot of the reverse-search, discount-model, or auction based systems haven’t worked well. Foxtons famously tried to expand to New York City using a discount model and failed spectacularly. ZipRealty IPO’ed back during the first dot-com era but has fallen to a very niche player at best. The failure of any discounting model gives more supporting evidence to the theory that there is extreme variability in broker skill. Maybe finding a home is high-stakes enough that no one wants a discount broker? Alternatively, maybe it’s a matter of survivor bias. Maybe lots of people try using the discount brokers, but they ultimately don’t transact, hence the seemingly small impact these models have on the market.

On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

There are two key differences that make real estate a bit harder to disintermediate than insurance. First, each home is a unique parcel of living space with infinitely many reasons a buyer might prefer home A over home B. Secondly, it’s usually considered essential to visit the new home prior to buying or signing a lease.

Health insurance, for example, has MANY finite parameters and options, but is still a largely quantifiable process of analyzing coverage, deductibles, and out-of-pocket limits. I applaud the Affordable Care Act for creating the bronze to platinum rating system, and hopefully someday market pressures completely commoditize plans into the “metal level” without concern for the underlying insurance company or network (the greatest difficulty would be the less quantifiable issue of whether enough of your preferred doctors are in-network).

To my knowledge, there is no concept of needing to physically visit or meet anyone prior to making an insurance decision, and in fact the human broker seems almost superfluous once all the insurance providers integrate with platforms like Zenefits or Gusto (which handles the administration, claims, and HR side of dealing with the insurance company).

On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

I agree that exhaustive inventory awareness is crucial, similar to how Li & Fung implied to customers that they had already done a comprehensive search of all possible suppliers, built the relationships necessary, and documented their capabilities. To maximize initial inventory, almost every site needs to partner with existing brokerages in order to bootstrap their supply. Without the supply, the severe network effect of a real estate marketplace nearly ensures failure.

Fortunately for all the companies in the space, real estate is notorious for low-cost multi-homing. Just as customers and drivers can choose to simultaneously appear on Uber, Lyft, and Sidecar; real estate suppliers can advertise on RentHop, but also a huge basket of competitors. As you mention, the network effect is even stronger, because unlike ride-sharing where each customer and driver is largely interchangeable, each parcel of real estate is unique.

The interesting implication is, when a large brokerage firm is advertising initially on an online website, who should pay who? The real estate firm argues that they are adding value to the site but sharing their inventory, therefore the Zillows of the world should pay for the data. On the other hand, the website operator argues they are spending vast amounts of marketing to ensure more leads, traffic, and business to the brokers, therefore the suppliers should pay to exist on the site. The answer is, it depends on the market dynamics, and more fragmented supplier markets are more likely to show the suppliers paying.

One last parting thought. Are all of the startups in this space initially agreeing to work with brokers in order to overcome the network effect problems, only to be secretly plotting a great back-door attack on the brokers? Once they are big enough to source supply and demand directly without the air of brokerage firms, is that the moment to pivot away into a pure distintermediation model? Zillow is the closest to achieving this goal, but with each earnings quarter, they announce even deeper partnerships and monetization models that include promoting the brokers and insist on the benefits of working with skilled agents.

On November 20, 2016, Lee commented on Why Real Estate Brokers Exist in 2016 And Beyond :

The unbundling approach is interesting and has been tried many times by startups (some dead, some still alive, but none who have hit a home run — even Redfin has pivoted away from discounting and unbundling). The idea is the 6% commission previously paid for the huge bundle of services we listed above, and yet “researching available inventory” might only be worth $250 to us while “negotiating the price” might be worth $10,000. Can we create an a la carte menu of broker services and allow the consumer to pick and choose? Presumably the unbundling also assigns a fair wage to the broker and eliminates the adverse selection component where brokers are only paid by the one customer who transacts and not by all the flakey customers who waste time without buying.

Some counterarguments for why that hasn’t worked is that a skilled broker is able to synergize all elements of the bundle to gather the best possible experience. For example, because the broker took you on all the open house tours, that same broker knows to update your search, or knows what levers to use during the negotiation. Or, for the seller, because the broker is getting paid the commission, they care more about exhausting all channels of marketing and more diligently follow up with any potential buyers that visited (in the best cases they mine their own rolodex for potential buyers).

I’ll leave with one final cool counterargument against the unbundling approach. Lets assign a fair hourly rate to real estate agents, say $25 an hour. What if the sum total of all labor content value done by real estate agents is actually LOWER than the sum of all commissions collected? I argue in some markets that is true, such as the New York City rental market. Thousands of agents spend time and marketing dollars showing apartments to tens of thousands of customers. However, less than 10% of all online inquiries seem to result in a closed deal, and less than 25% of physical showings lead to a closed deal. It is possible that the average of all agents LOSE money, but because new agents enter the system all the time and below average agents quit the industry in disgust, then there are always a fresh batch of agents willing to work. If we believed the industry was a net-loss to agents but subsidized through recruiting and attrition, then any unbundling and a la carte approaches to broker services should be doomed to fail.

Many of you have convinced me to rethink the harsh cries for an immediate end to the subsidy. After reading through comments, the primary takeaway is to try to research the true elasticity of demand, and in particular stay mindful of likely substitution effects to competing cars. I have always found the range of Tesla buyers to be an intriguing mix of three different categories: the car enthusiast, the luxury environmentalist, and the tech geek (autopilot hobbyists fall into this category). Speaking as a member of the final group, I had thought there was no substitution effect on demand because there is no other way for a consumer to purchase a self-driving car. However, you’ve all convinced me that the generic fast-car, “Ludicrous Mode” enthusiast does have other viable options in which $7500 can tip the scales.

On November 7, 2016, Lee commented on No, my name is not like the baseball equipment brand :

Wow, I was in Seattle cheering for the Mariners during Ichiro’s amazing reign there, and then had the pleasure of moving to NYC just as he joined the Yankees. I suppose I never noticed his bat because most of the action comes from watching him steal bases! I am impressed that professional sports take the extra step to worry about environmental issues. To be honest, the MLB is one area where I would have been willing to give them a free pass. It is, after all, a world class competition and when there are so many fans and so much brand value on the line, I would expect everyone to do everything they can to gain every shred of advantage. I’m curious, if there were no rules or restrictions on material, would the rare wooden bat rank comparably to the best state-of-the-art bats?

On November 7, 2016, Lee commented on PetroChina: Can Deep Pockets Reach Deep Shale? :

I am very happy to see more major players entering the crude oil markets, especially one that would absolutely not be intimidated by the existent cartels and pseudo-cartels. The events of the past 3 weeks are a case in point that commodities markets are much more volatile than necessary as a result of geopolitical jockeying by players with differing motives. In the past 2 years, we have seen front-month crude oil prices collapse from over $100/barrel down to a low of mid $20s earlier this year. When OPEC announced an agreement to cut production three weeks ago, we finally seemed to break past the resistance point of $51, only to have Russia quash the rally by saying they would not play along (down 20% as of last Friday).

Markets should on net ideally reduce the volatility, and instead the government policies and movements seem designed to manipulate markets instead. I would say the more players the better: more liquidity, more volume, and hopefully more transparency.

On November 7, 2016, Lee commented on Climate change… Understanding the meat of the problem :

I find the food tech industry extremely interesting and I can’t believe more work isn’t being done to research ways to innovate all aspects of food production and preparation. Some might point to the seemingly difficult upfront capital and research costs (just like in our Indigo case), but I am actually more amazed there hasn’t been more work done in engineered nutrition. Why is processed food almost always less healthy AND less tasty than the natural version? Humans have figured out how to build incredible tools, computers, machines, medicine, and chemical compounds. Food has been around since the very beginning of time. Why have we had almost no innovation in food processing beyond cooking and preserving?

You bring up a great point that climate change is causing somewhat unpredictable weather extremes, which can skew perceptions about whether climate change is even a problem. To your point regarding the record snowfall in New York City, a certainly lived in Manhattan through all of that snow, and I distinctly remember comments such as “wow, it’s so cold and snowy, we don’t need to worry at all about global warming!” It appears to be largely an awareness and education issue, and while trends are quite clear in the long run, short term fluctuations can be very misleading.

On November 7, 2016, Lee commented on Insuring Against Climate Change :

Very interesting that we are starting to see climate change factor into quantitative modeling for risk and insurance purposes! Many years back during my hedge fund days, I asked the energy and commodities traders whether they could see any evidence of climate change creeping into their forecasting models or in any market prices. At the time, it was a firm “No”, mostly because the impact was far too small relative to the daily volatility of the asset class. Also, most commodities futures are only liquid enough about one year to expiry. I could imagine a not-so-distant future where climate change creeps into the models; a slight but necessary and important adjustment, similar to the way we build in convexity corrections into the long end of a fixed-income yield curve.