16 Mar 2018

Cryptocurrencies in one figure

Pretty cool:

The bliss of disconnection

Gaucho with a smart phone
I've felt a lot more relaxed since I quit Facebook,* shut down several twitter accounts, and deleted all the "traffic loggers" that I had running on various sites. I have fewer updates to track, no "missed messages" to regret, and more time to think without pings, alerts and other cues pushing me to be a good social animal.

I suggest that you look into the same type of action, if you're feeling stressed and behind all the time.

As positive side effects, you will have better conversations with friends and family (assuming they can also "put down that app"), more time for actions you pull to yourself (rather than having actions pushed to you in the "feed"), and perhaps even more time to comment here!

When we were lost in Uruguay a few summers ago, we stopped to ask a gaucho for directions. He said "no hay GPS-ay?" and I said no. We had an amusing conversation in my bad Spanish, and he even introduced me to his son, by phone, who was studying in London. It was a nice side effect of lacking GPS.

Bottom line: Take the time to talk with people (or write a decent email). It will let you focus and build clearer communications with other people.
* A consultant told me this morning how "login with Facebook for free wifi" hotspots grab your IMEI code (off your phone) and then track you in the store and all other spots where you might walk (without using wifi), so those stores can target ads at you based on the your Facebook profile (did you read all the details of "I agree"?)

15 Mar 2018

The "plastic-free market" is no big deal

A few weeks ago, there was a huge small wave of news reports on the "plastic free" aisle installed in a Dutch health foods store.

That store (Ekoplaza) happens to be 5 minutes away from my house, so I rode by to check it out.

As you can see, the "aisle" (in what used to be a bike parking garage) is just a collection of foods packaged in glass, "bio-plastic" or nothing at all. That's not revolutionary.

What's worse, the "plastic free aisle" also has a "Bluewater" kiosk outside that is supposed to reduce the use of single-serve plastic bottles by giving you a place to fill your reusable (plastic or metal) bottle. Although this is a nice idea, I think that the kiosk probably represents a negative environmental impact, given Amsterdam's excellent tap water quality. It seems to be there as a giant advertisement for "buy our water filtering product." Fail. Update: It's gone! Yay!

Bottom line: The best way to reduce plastic use is to tax its source (oil, via a carbon tax). The best way to reduce pollution from plastic water bottles is to add deposits to their price, to pay for recycling and re-collection. After these simple actions, then you need to consider the total costs/benefits of switching to plastic from paper or glass. In many cases, plastic has a smaller environmental footprint than its "virtuous" replacement.

14 Mar 2018

Links of interest

  1. "If it’s unclear why modern architecture must be so offensive, it’s particularly puzzling why French architecture must be"
  2. The Oxford English Dictionary in the internet age
  3. John Perry Barlow on why it's important to talk about using LSD (I do sometimes...) and how cyberspace can bring peace ("direct communication should breed understanding and tolerance..." but some people didn't get this memo). Here's a fascinating 1992 discussion of cryptography and free speech rights.
  4. Yay! "Emissions trading now covers 15% of global emissions"
  5. Carbon capture might cost as little as 0.10 percent of world GDP. We need to get going!
  6. A Louisiana mayor "invests" millions to force government to rescue the town from rising seas (sunk cost fallacy in action)
  7. Capitalist geeks visit Eastern Europe around 1990. Much has changed.
  8. The OECD has published its "Principles on water governance: from policy standards to practice." Corruption is listed as "something to watch" so I think this advice is a waste of time in the many countries where governance failure arises from corruption.
  9. Dutch cities are going to put more deposits on plastic and aluminum containers, especially after the industry's claims of "not cost effective" turned out to be lies.
  10. Worried about Fake News and too much noise on social media? Read the paper. (Speaking of quality reading, JSTOR has daily newsletters!)
  11. Listen to a real CEO talk about running businesses and managing people
H/T to PH

13 Mar 2018

Is Lending Club a failure -- or fraud?

When I heard about Lending Club, I thought it was a brilliant idea: a market that would allow investors to provide loans at low risk to borrowers who wanted to pay less interest than they would normally do at a bank or via credit cards.

The key attraction for me as an investor was that I would be bear less risk because I would only be lending $25 to each borrower, giving me a diversified portfolio of many small loans. Diversification is a time-tested strategy for lowering risk while protecting returns because it allows you to get a decent average return without facing a large catastrophic loss.

As an example, compare the uncertainty of making one $5,000 loan to a borrower with the risk of making 200 $25 loans to 200 borrowers. In the case of many borrowers, each with a 5 percent chance of default, there is an expected loss of 10 loans, or $250 in principal. In the case of a single borrower, you cannot use a risk assumption, as it's either default or not (losing $5000 or nothing), which is far less attractive to risk averse investors like me.

Lending Club was exciting because their platform would make it easy for investors like me to match with thousands of borrowers. All Lending Club needed to do was group borrowers by risk (into "grades") and price that risk correctly (the interest rate). My net return would depend on how much interest I earned against how much I lost to defaults.

The statistics for loans issued in 2012 show interest rates and defaults for different grades:

As you can see, it's possible to get an "Adjusted Net Annualized Return" (ANAR) of 6-9 percent with a portfolio of lower grade loans (B-G), so losses should not be a problem. Indeed, one commentator said exactly this in 2011:
Q: What are the risks of investing in peer to peer lending sites like Lending Club or Prosper.com?

A: Lending Club and Prosper both have analyst that does risk management. If you blindly invest and be well diversified, the returns will beat any money market/CD today. The biggest risk in my opinion is the change in underwriting/model that Lending club or Prosper do to increase their volume and adjust to compete. Yes, defaults risk exists, and they are factored into the interest already by Lending Club and Prosper.
Another commentator in the same thread said that risk arose from borrower default but also "Unreliable forecasts of ROI and default rates; this did apply to Prosper lenders in the past. If you did lend in 2006/2007 and believed the forecasts you might be hard hit; actual default rates for loans were as high as 30 percent."

Although I understood those ideas and risks, I trusted that Lending Club was going to do a decent job at underwriting (deciding how risky borrowers were) and thus pricing risk in its interest rates.

I was wrong on that, and my returns have been so poor that I think that Lending Club may be guilty of failure (poor service) and perhaps fraud (misrepresenting its product).

Here's my story:
  • I deposited $65k into two different Lending Club accounts ($25k into investment and $40k into IRA) in 2015.
  • In both accounts, default losses started to exceed interest income after 12 months.
  • According to Lending Club, I should not be seeing such a high rate of losses, as I was diversified across ~3,000 loans in two unrelated portfolios.
Was I taking too much risk? My combined portfolios had 8 percent A-grade, 57 percent E-grade, and 44 percent F&G grade loans. I was certainly being aggressive, but not overly so if you look at the 2012 figures available when I put in my money. Using the numbers above, my weighted return should have been 9.87 percent.

My actual experience was much worse (note the 7-9 percent drop in E/F/G grades)

Using these numbers, my weighted return should be around 1.52 percent. By my calculation, it is -0.44 percent at the moment (and falling each month because losses from defaults are greater than the interest I am receiving).

Note that my method of calculating returns is slightly different from Lending Club's ANAR. I am comparing net cash (withdrawals-deposits), but Lending Club assumes you reinvest repayments into new loans, which boosts your returns in the short run (payments are more likely in the first months after a loan is issued) at a cost of keeping you on their platform longer (since you always need to wait 3 or 5 years to get your principal back). Such a strategy makes sense if Lending Club has accurate figures for default (and thus sets the interest rate correctly), but it is asking you to throw good money after bad if Lending Club's model is biased or fraudulent (more below).

When I first sent money to Lending Club, I had no idea that I would have to worry about this issue, as I assumed that diversification would help me achieve the returns that Lending Club advertised. Sadly, I was soon to learn that the standard investment caveat ("past returns are no guarantee of future performance") applied to me. It shouldn't have with loans, but it did, for the same reasons that it did in the 2007-8 financial crisis: poor or fraudulent underwriting.

The following figures of my net gains on each of my account illustrate my learning process:

Net gain (interest less losses) for investment account
Net gain (interest less losses) for IRA account (Note $20k added just before the other account started losing)
So you can see in these figures that my upward trajectory slowed, flattened and then plunged as a growing share of borrowers stopped paying back their loans. (Note that the plunge was not because I started selling on the secondary market, as those sales should have slowed my losses by reducing my portfolio size.)

Although I was hopeful that "the bad ones" would drop out early, leaving only good loans, I decided that hope was a terrible strategy, as it's more likely that defaults would increase as the loans matured. I therefore started to sell loans on foliofn, Lending Club's a per-to-peer loan secondary market. My hope was that I would cash out of loans at their face value before borrowers defaulted, thereby transferring my risk to other investors. I was able to sell many loans in this way, but I had to discount some of them and could not sell others. At the moment, I have about $3k left with Lending Club, which is a small enough amount for me to let them mature or default without causing me too much worry.

For the 36-month loans that I did not sell, I lost more money to defaults ($5,100) than I made from interest ($2,100). That's not a good investment.

So if you've made it this far, you will realize that I was wrong to assume that I was going to make as much as earlier investors in Lending Club. That conclusion leads to two possible explanations: either I was unlucky with my "diversification" across 3,000+ borrowers (in a growing economy!) or Lending Club's business model is biased or fraudulent.

Let's look at unlucky diversification, i.e., my returns vs the average for "accounts like mine" (the blue dot below the mass of dots is me) for my two different (uncorrelated!) accounts:

Fool me once, shame on you.

Fool me twice, shame on me!

These low returns can have two possible explanations. Option one (failure): the automatic investing bot was extraordinarily unlucky in choosing my 3,000+ different loans in two different portfolios. Option two (fraud): The reference figures are wrong (lies) and my "bad" returns are actually average returns.

Either way, I'm screwed and Lending Club is not -- and that's the key.

I realized (too late) that there was a problem with incentives. Let me explain using some economic jargon. In a principal-agent relationship, the principal pays the agent to represent the principal's interests. The risk to the principal is that the agent is either incompetent at their job (adverse selection) or uninterested in working hard (moral hazard).

The irony of this whole post is that I teach these ideas to students. Look how clever I am!

In the subprime crisis many banks, underwriters and credit rating agencies (CRAs) gave unqualified borrowers loans with the knowledge that those loans would be sold to investors who would trust the over-optimistic credit ratings given by the banks/CRAs. The bankers didn't care (moral hazard) because they made their money from issuing the loans and did not carry any risk if the borrower defaulted.

The situation with Lending Club is exactly parallel: Lending Club earns fees from making loans that are sold to investors like me. Lending Club has no "skin in the game" if those loans go bad. Investors, not Lending Club, lose money if Lending Club cannot or will not do its job correctly.

Just to double check, I called Lending Club and asked "what do you guys lose if borrowers default on me?" The customer service representative replied "Nothing. Sir, we look at your borrowers very carefully."

Right. Let's look at some of these "carefully vetted" borrowers who defaulted on my loans:

This borrower made zero payments on the loan.

This borrower made one payment (2.4% of my $25 back) and went bankrupt 5 months later.
As soon as I realized that the game was rigged, I started to look around for other opinions and experiences with Lending Club. I was not pleased with what I found:
Wole on Lending Memo (Sep 2016): "I also do think defaults have gone up. I'm also having a harder time getting quality notes to investing in the C-E range... I'm thinking that's due to the flow in of institutional cash into the platform."

Mr Money Mustache (Oct 2016): "I saw my account balance drop for the first time. The balance went down $300, when statistically it should be up about $600 over that time period. With my balance diversified across over 2,500 notes, this should be a highly unlikely event."

A data scientist (Nov 2016): "It's much more of a clever hack than an actual, sustainable, lasting, value-creating enterprise. One of the biggest flagship FinTech companies, Lending Club, is in a ton of trouble."

Mike on Lending Memo (Jan 2017): "Looking at the 4 year trend, LC is a joke and it is only a matter of time before it implodes. No bank or lending institution could keep their doors open with statistics like this... Somebody is making money, but it would appear it will not be the average investor..."

Mike on Lending Memo (Jan 2017): "As I have said in the past, as people learn that LC has no teeth in recovery... As more learn they can borrow and not repay and get a slap on the wrist from their credit score, I wouldn’t want further exposure in this platform for that reason alone."

John on Lending Memo (Feb 2017): "A friend on a paid account with not much more overall investment has seen much better returns (7% or so adjusted vs. my 2%). It makes sense that services like LR and NSRInvest [sites like that help you choose loans on Lending Club] weigh the quality of the loans to their paid and largest investors who pay their bills. [snip] And now LR has introduced ‘LR Series’ aimed at accredited investors, with an updated algorithm and very large minimum investment. This trickles down -- by its nature worse and worse loans are available to the people at the bottom of the food chain... True P2P Lending via their marketplace is a very small part of Lending Club’s business -- they gear their services to these large investors."
From these comments (and my earlier analysis), we can get a better idea of what's happened:
  1. Lending Club started off well, checking borrowers for their creditworthiness, setting appropriate rates, and selling loans to retail investors.
  2. This success attracted more borrowers and investors, which put more stress on the platform.
  3. It also attracted institutional investors who wanted better risk management and a first choice of loans.
  4. Not knowing that underwriting quality had deteriorated and that loans were being cherry picked, new retail investors like me got mis-priced over-risky loans.
  5. Lending Club knows we are losing money but doesn't care, because "it's our risk," and they are making the big money from institutional investors.
Bottom line: Lending Club is no longer a decent investment, just like sub-prime mortagages were a bad investment in 2007. If you're in, get out. If you're thinking about Lending Club, keep your money in your 0.2 percent interest savings account. You'll make more than I did.
Addendum: A reaction from Mike (the same as above) on this post:
David. Hope people listen and not keep being taken with this scam. You could make decent returns similar to their estimates 3-4 years ago. When they first played with their underwriting rules is when things started going downhill, which I believe coincided with opening up to institutional investors. Was a great platform before that happened. I’m still close to 5% with only about 6K left to age out, but still loosing monthly at this point. Even though I stuck to 36 month loans the last couple years thinking that would help, certainly has not been the case. Even the folks on this thread that would not listen. Wonder if they are kicking themselves yet? Thanks again for the article.

Bottom line, there is no reward without risk. Stock market is unpredictable, but the rewards are far better, and their are tools and professional periodicals to help manage your money wisely. Also, if there is a buck to be made, be sure somebody with real money will plug the hole for the common man to benefit.

So made 20% last year in the stock market and -4-5% in Lending Club, and can’t even tell you what my loss is because Lending Club isn’t willing to show you. Anybody that reads this thread. Stay away from LC, unless you like throwing your money away.

Sorry for your losses.

12 Mar 2018

What's your insight on water management?

For my birthday (last August), I asked people on my mailing list for their most important insight or advice on water (or environmental) management. Here are their replies:
  • "In the end, sustaining and rebuilding natural resilience may be our only real defence against the ravages of climate change." -- Ralph Pentland
  • "One Water" -- Celeste CantĂș
  • "All the governance in the world won’t help you if you don’t know how much water is in your basin…" -- Sarah Hendry
  • "As one who grew up in the pretty much desert - we water our gardens or the grass (or agricultural fields) in the early hours of the morning, or after sunset, since water loss due to evaporation decreases." -- Keren Cohen
It's a bit late (or early?) for my birthday, but what is your insight into water management?

Monday funnies!

This cartoon (click here for full size) is from r/PolandBall.

By coincidence, Trump has also inspired me to coin a new hashtag, #mega, for Make Emigration Great Again. Consider yourself invited to the socialist, failed old world.

10 Mar 2018

Houden wij van natuurijs? Do we love natural ice?

This Dutchie is skating in honor of natural ice, the ice that defines Dutch culture in the countryside and in the cities. She is donating money to protect Dutch natural heritage, a step that's very helpful in the face of climate change. I'm a miserable skater but sent them money too.

H/T to TR

9 Mar 2018

Academic Just So stories

Sometimes academics present a "just so" version of history that lays aside complexities interfering with their narrative and hypothesis.

This habit is not unique to academics, as humans love to tell stories (reason reason reason)

  • Guns Germs and Steel is a wonderful book, but some historians question its scope and conclusions
  • The author of Shadow Wars claims that Western merchantilists are the "main" forces affecting countries around the world, but such a claim overlooks the role of other powers (Russia, Japan, China) as well as local actors (hello corruption!)
  • Was Europe's "rise" based on separate nation states that shared intellectual developments, or perhaps their struggles to beat each other and then take over the world in the 17th to 20th centuries?
Bottom Line: Human social, political and economic interactions are far too complicated to fit into a clean narrative, so question just-so stories and look for other important influences if you want to avoid an over-simplified perspective on reality.