Hacker News Reader

Comparing the LIBOR spread isn't enough to compare senior subprime tranches to senior prime tranches. Among other things, the attachment point matters a very a great deal.

The attachment point is the % of cumulative losses after which a given tranche starts suffering losses of principal. So for example:

L+20 with a 20% attachment point against historical max losses in the asset class over all credit cycles of 1% is a lot better on a risk adjusted basis than L+200 with a 40% attachment point against historical max loss experience of 15%.

Without talking about historical loss experience we can't really guesstimate the margin of safety here nor say that a given tranche is or isn't good value.

Another possibly interesting nitpick: LIBOR isn't risk-free. It's an interbank rate so it is the short-term yields paid by highly rated financial institutions. Risk free means backed by an entity that can print money, like the Fed. That's why the LIBOR-Fed Funds basis exists.

Also, re: "I think it's interesting how high interest rate risky loans can be considered as sleazy while giving opportunities to the less fortunate is considered an admirable goal." check out the book _Scarcity_ by Mullanaithan and Shafir, which has a super interesting take on this question!

Completely agree with everything you said. I was only calling libor risk-free as it is what would be substituted for the risk free rate used in finance textbooks. Technically it's not risk-free but serves as a base return to benchmark off of. Also, most of these bonds pay coupons linked to libor.

Attachment and detachment point are definitely very important. I found a random fed document from 2012 that suggests similar attachment/detachment points for AAA prime vs subprime abs securities (~79% attachment). I haven't looked at it too closely though and I could be missing something. I know spreads are a lot tighter since 2012 and credit quality of underlying loans may have worsened with lower standards. On the other hand, people are a lot better off now that in 2012 (i.e. stock market, housing market, etc).

Will check out Scarcity. Thanks for the recommendation!

[1] http://www.federalreserve.gov/SECRS/2012/May/20120523/R-1401...

Current state of economic thinking: technically not risk-free, but we'll call it that anyway...what could go wrong!?

Well, if you're talking about LIBOR, the risk is that the AA bank to whom you have counterparty risk will default within 3mo, so it's very, very, very low risk.

> The reason there is such appetite for these products is because of the low interest rate environment. Investment managers are desperate to earn a slightly higher return that the risk free rate. This is driving demand which explains why you saw the issuance of subprime abs explode over the last year and spreads tighten (spreads have widened later in the year though). Credit standards have probably degraded to try and meet the demand of abs securities, but that's inevitable.

This part terrifies me. Not because of auto loans, but as returns remain low (and in my opinion, structural changes have occurred that will keep rates low in perpetuity), investment money will continue to look for higher returns with no moral qualms. Like farm land:

http://www.forbes.com/sites/joshuarogers/2014/09/23/dirt-che...

Its easy to look past subprime auto lending and payday loan lenders, but what happens when natural resources necessary for survival (food and water in this case) are sold to the highest bidder?

Perpetually low interest rates are a big concern to pretty much everyone in the industry apart from those employed at the Fed. Europe just cut their rates to negative. That means that investors have to pay countries to hold their money for them (I'll lend you $100, pay me back $99 in a few years).

I loved the (very defensive) Bloomberg headline yesterday: Owners of Negative Yielding Sovereign Bonds Say They're No Fools [0]

The idea is that every country is trying to devalue their currency in order to help exports and domestic producers stay competitive. The countries are newly emboldened by the fact that rates have been very low for very long without much inflation. But economics never works exactly like its described in textbooks. I fear that if this all explodes we could have high inflation with a stagnant economy with no magic levers to pull to address both problems. In that case, it will be about survival.

[0] http://www.bloomberg.com/news/print/2015-01-26/owners-of-neg...

> The idea is that every country is trying to devalue their currency in order to help exports and domestic producers stay competitive. The countries are newly emboldened by the fact that rates have been very low for very long without much inflation. But economics never works exactly like its described in textbooks. I fear that if this all explodes we could have high inflation with a stagnant economy with no magic levers to pull to address both problems. In that case, it will be about survival.

I'm glad I'm not the only one who thinks about this.

I don't understand why you have concerns of high inflation. This has not been a threat and isn't a threat on the horizon. If it was, that would be a good thing. Maybe they're not "magic levers" but there are tools that would help with that.

We already have low (consistently below target) inflation and stagnation.

The idea that we are going to get to a place with high inflation without climbing out of stagnation on some level must be explained, and it would be interesting to see how this would happen in your view.

If economics never functions like it does in the textbooks, then is macroeconomics a random activity? Yet it is predictable to the point where you expect runaway inflation?

There has been past events of inflation without climbing out of stagnation. It's called stagflation [0] which occurred in late 70s in the US.

I don't believe the macro economy is random. But I don't believe it can be understood to any certainty worthy of the quantitative methods ascribed to it by modern economists. If you look at empirical evidence (regression) for what most consider economic truths, you'll see a shockingly low R^2.

That being said, I think some economic truths can be discovered through logic. For instance, all else being equal, an increase in money supply will result in higher prices. As Milton Friedman said: "Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output".

I can't think of an example of destructive inflation without the increase of money supply by a central authority. So certainly there must be some link. So maybe we can say inflation is (almost) always associated with an increase in money supply but an increase in money supply is not always followed by inflation.

I don't know why it hasn't happened in this case.

[0] http://en.wikipedia.org/wiki/Stagflation

I think part of the reason is that history always seems compressed when you look back, but the present seems to stretch forever.

So you can look back at a 5-7 year period in the 1970s and say 'look, that's where inflation occurred'. But living through a seven year period feels like a long time.

The final point is that inflation statistics are calculated in different ways now. If you look at inflation measures not linked to a 'basket of goods' type measure, there is plenty to see. The breakouts in collectible markets in the last year or two certainly point to strong inflation.

But on the whole the government response is not as stupid as it was in the 1970s, when they did things like put in price controls and things like that. There was a big deflationary hole to fill in first.

I don't think we are anywhere near out of the woods yet.

Interest rates should trend to 0 (and they have). There's not much, if any, need for rent seekers. That rates have been able to be sustained so far is more of a historical accident, propped up by intentional exploitation of the productive classes.

How about a system that doesn't rely on opaquely modifying the global supply of fiat currency while simultaneously tying the destruction of the created money to an exponential penalty curve where you're essentially required to scrap around the planet performing services on demand for others who contracted under other penalty curves, lest their 'investors' get 'hurt'?

Right - the overall impression I got of ether article is the last big deal was $1 billion. which is peanuts in the scheme of things, really.

Some people might get burnt, but it's not going to crater the economy.

Subprime has become a dirty word by association, but it really just does mean lending at a higher rate to people not as credit worthy. Which is a huge chunk of the market.

bko, sorry to be OT, but where did you learn about finance?

Undergrad and MBA but I barely learned anything about structured products through any of my classes. Its mostly through working in the industry. It's very hard to find good data, especially market data for these more esoteric products (spreads, yields, assumptions). A lot of companies make a lot of money selling this data (I mean A LOT). For instance, a weekly financial newsletter on asset backed securities (~10 pages long, pdf, emailed out) is $3,500 a year [0]. It's a good read though and definitely contains info you can't find in other media outlets

Anyway, if you're interested in structured products, I would just google '[structured product] primer'. There are a lot of great primers publicly available. Most are really old (2006 - present) but don't let that put you off. All the fundamentals are the same and they're definitely worth a read. Some are more relevant since they're still being structured. Examples include ABS (which include prime and subprime auto) and CLOs (collateralized loans). Others (like non-agency, subprime, most CDOs) are not. This [1] book was also helpful and more in-depth than most primers.

Feel free to pm me for more info.

[0] https://www.abalert.com/subscriber/subscribe_now.pl [1] http://www.wiley.com/WileyCDA/WileyTitle/productCd-111800674...

Thanks for the info, will look into the Wiley book. HN doesn't have private messages, but you can leave your email in the profile section.