This is an explanation of the cause of the financial crisis, though it’s too long at 17 pages. Here’s my dumbed down summary. Around the world there was a huge glut of cash, particularly in Asia. Individuals can put cash in FDIC secured savings account, but no such safe account exists for huge deposits. So where can money markets and hedge funds and Chinese banks put their money and guarantee it’s safe? They put it in the repo market.
A money market fund (MMF) has $1B in cash and no safe place to deposit it. The investment banks came up with an idea: give them $1B cash and they’ll give the MMF $1B in Aaa-rated Treasury bonds as collateral. But the MMF wants to be able to withdraw money at anytime, so the bank agrees to buy back those bonds the very next day. This is called a repurchase agreement and is the foundation of the repo market, worth over $12 trillion worldwide. From the MMF’s point of view, they earn interest on their cash and hold an equivalent amount of bonds just in case the bank fails. And they get their cash back every day, so they can withdraw what they need in the morning and sign another repo agreement for the next day. From the bank’s point of view, the repo market turns their vast holding of Aaa bonds (required for all sorts of purposes) into cash they can use to do more lucrative investments. Everyone is happy.
The problem lies in those Aaa-rated bonds. The repo market got super huge and there wasn’t enough high quality bonds to meet demand. Someone figured out how to turn consumer loans (home, credit cards, auto, school) into Aaa-rated bonds. The bank gives the MMF $1B in consumer loans, and AIG promises to make up any loss in those bonds. AIG did this because everyone agreed that consumer loans would never blow up, so they earned a small fee on every loan for doing nothing. AIG made tons of money. Everyone is happy.
In 2007 home values suddenly declined a bit for the first time ever. The MMF complains the $1B in mortgage bonds have lost value, so they want more bonds as collateral or their cash back. All the bank’s cash is tied up in other investments, so they have to sell some stuff to raise money for the MMF. But all the banks started selling the same securities, which sent prices down. The banks suffered losses and had to sell more securities to make up for their losses, which sent prices down again in a terrible cycle. This initiated the banking crisis around Sept. 2007. The global economy crumbles. Everyone is freaked out.
The paper says that this was a classic run on the bank; that is, suddenly everyone wanted their money back at the same time. FDIC insurance stopped bank runs by consumers, but the lack of insurance for large depositors means we’ll continue to have this same crisis in the near future. Somehow the banking system must be made more resilient to panics. I don’t know if this is a complete explanation of the financial crisis, but it does help to understand all this weird “parallel banking system”.
This article (and this one) pokes a hole in the idea that there’s a sustainable first-mover advantage for businesses. That is, being the first search engine or social networking service did not give those innovators any advantage in the market; in fact, most of those first generation companies are gone (Excite & SixDegrees). This strikes me as obvious, yet VCs insist that being first to market is critically important. Has Microsoft ever built the first of anything? Is the iPhone the first ever smartphone? Even Amazon came several years after Book Stacks Unlimited. There are few successful tech companies that were the original pioneer in that market. Most were second (or third or fourth) movers. How many search engines came before Google? They couldn’t get much funding because that market was seen as a dead end.
The problem with being a first mover is (1) you have to create everything yourself and (2) you have to convince customers to take a chance on your crazy new product. These are especially difficult if you are a startup. The problem with (1) is it takes a lot of trial-and-error to come up with something. You may have to invent some tech to make it work. You may rely on tech or standards that ultimately limit what you can do. The issue with (2) is that you spend all your time and money convincing Innovators and Early Adopters, and then someone else jumps in right when the Early Majority is ready to buy. The second mover will save time and money on (1) because you’ve done all the work. And they free ride on your evangelism and enter the market after you’ve generated demand.
So the right tactic appears to be to jump into a market that the early adopters are excited about, but that isn’t implemented all that well and hasn’t hit the mainstream. I think Twitter is ripe for a second-mover takeover. They’ve validated the idea of micro-blogging, but there is still tons of innovations that can be layered on top. Twitter won’t be able to shift direction too quickly without upsetting their established user base. And their fragile infrastructure also makes it difficult to change. Now if only I had an idea…
The NYTimes has finally decided to move to a subscription model for their website, modeled on the Financial Times. This article does not mention an interesting nugget that was in the memo to the NYTimes staff: “There has also been much speculation in the media and elsewhere about whether The Times will join a consortium as part of the metered model implementation plan. At this stage, our plan is to introduce the metered model as a stand-alone product. At the same time, we continue to discuss alternatives with a broad range of prospective collaborators with regard to bundled offers and other aggregation opportunities.” As I suggested earlier, more sites should join a potential NYTimes consortium. For example, I can see The New Yorker, The Atlantic Monthly, and other magazines that appeal to that customer segment easily joining a consortium. That makes a subscription more appealing and kicks a few extra bucks to those magazines, too. Now all they need to do is implement my suggestion for a slimmer dead tree version, preferably a competitor to USA Today.
Yesterday, a few Wall St. CEOs faced a wave of stupid questions from the Financial Crisis Inquiry Commission. Jamie Dimon, head of JPMorganChase, made a flippant but important comment: financial crises happen “every five to seven years.” There’s even a book out called This Time is Different: Eight Centuries of Financial Folly detailing the long history of booms and busts. Warren Buffett has said the boom/bust cycle will continue because fear and greed are hard-wired into human beings. Whatever rules the gov’t throws together today will fix the last problem (housing) but do nothing for the next problem. Also, these rules will be watered down by lobbyists, and further eroded over time as the public forgets and Congress is bribed by their Wall St. masters.
Booms & busts are inevitable. When a bubble pops, the gov’t will need to spend lots of money to prevent the economy from falling into a depression. The solution is to pass a Tobin Tax, a tiny tax on financial transactions. This tax can be designed to be progressive, ranging from 0.1% to 0.5%. If you buy $1M worth of stock, you pay $5K in taxes. At the lower tax rate, $10K in stock would cost $10 in taxes. It can be tweaked to catch most cheaters. Furthermore, the SEC should push most OTC trading to exchanges, which further increases the funds generated.
The money raised by this Tobin Tax should be reserved for the inevitable busts that require gov’t action. Just like a national disaster, the Fed would declare a financial disaster and use the funds to prevent a depression. During good years, the money will be used to pay off the last bubble and save money for the next. Along with authority to take over failing banks and smoothly enter bankruptcy, a financial disaster fund will cure the hangover after a bubble.
[Obama is on TV right now proposing an excise tax on banks to recoup $100B probably lost from TARP. It’s fine, but the funny business will just move to hedge funds. Bankers are like cockroaches, you need to gas the whole place or they’ll keep coming back.]
I recently signed up for a Netflix trial account and rated around 300 movies. I noticed that I was giving out lots of 3 and 4 stars, a fair number of 5s, but very few 1s and 2s. It reminded me of this article, which says the average rating for products and movies, at lots of different sites, is around 4.3 out of 5. Here’s a post from Youtube showing that most ratings are 5s. Another from Yelp. The article asserts that people are too nice or, for some inexplicable reason, currying favor (from whom? why?). The implications of this are interesting for any collaborative filtering, or “crowdsourcing”, algorithms because if ratings are skewed high than you need to factor that into your recommendation system.
Since I am not a nice person nor am I sucking up to faceless corporate entities, why are my ratings generally favorable? The reason is that I’ve already filtered out the movies I know I’m going to hate by simply not ever watching them. I hate chick flicks, most goofball comedies, most horror movies, all musicals, many action movies and most epic films. Since I won’t watch any of these movies in the first place, I can’t give them the 1 star rating they deserve. The few times I really hate a movie is when I go simply because the movie is popular. I hate all the Harry Potter movies, but I watch them because everyone else seems to like them. (You are all idiots!) Same goes for books and products and restaurants. I don’t go to random restaurants hoping for a good meal. I read the reviews on Yelp and Chowhound and check their menus online before walking in the door. I buy electronic products after reading the reviews on Newegg and studying the specs. I read reviews and watch interviews with the authors before picking up a book. Most people use a similar suite of filters to pick a movie, book, or product; therefore, they will generally be pleased with what they get. A user’s average rating should be much higher than 3 stars (average). In fact, a user’s average rating is really a measure of his filtering scheme: the better his filters the more pleased he will be with his final choices.
Since the average ratings will be skewed too high, how can you get useful information from collaborative rating systems? For Netflix, they should more aggressively ask if you are even willing to see a particular movie (yes/no) and feed that info back into the recommendation system. The Netflix dataset does not contain this information. Yelp could do the same: does this restaurant sound interesting (yes/no)? For Amazon, it might help to allow customers to save a few choices so they can do a side-by-side comparison. For example, I could select 5 40” LCD TVs that meet my requirements. Amazon could have a slick interface that helps me compare and eliminate them one-by-one, leaving me with my final choice which implicitly gets a high rating because I chose to buy it. Apparently, eBay has a good solution of primarily looking at the percentage of negative sellers’ ratings because those are angry customers. Too many angry customers and you are booted from the site. All these systems can be improved by determining which choices a customer discarded on the way to making a final, usually satisfying, decision.
A new workshop at OOPSLA is aimed at finding ways to evaluate the effectiveness of programming languages and tools. Unfortunately, the entire conference rests on a single problem: how to measure programmer productivity. Except for pointy-haired bosses, most software professionals believe productivity can not be accurately measured. This workshop will likely be yet another outlet for HCI research with awful methodologies.
Someone told me a few months ago that Orbitz will detect and raise prices for some customers. It didn’t occur to me to check this until last night. I found a good deal on Orbitz for a hotel+car package, but looked around the web for better deals. After a few minutes, I reran the search on Orbitz and the same deal came up $200 more expensive. I fired up IE’s InPrivate browser and tried again. Now the original, cheaper price showed up. I reran the search in a previous browser and still got the $200 extra charge. Because I lack an MBA, I don’t understand how screwing your frequent customers is a good business tactic. I’ll be searching Orbitz and other travel sites using InPrivate or InFilter mode from now on because I don’t trust their prices anymore. If more people find out about this, people will lose trust in Orbitz and use it less frequently. I suggest people use a different browser or computer to verify that you aren’t getting screwed on prices, particularly package deals.