Hacker News Reader

I think the issue was the collusion in making up this number, and not the fact that the number was made up.

You're right about every single thing you said. However, the conclusion is slightly off.

Think about prices at a gas station. Each gas station owner is free to price their gas at a price at which he thinks people will buy it from him. This is normal and most definitely legal.

What is not legal is if even two or more has station owners make a secret pact to keep prices the same. This is called price fixing. It does not matter if the price is a bit higher than normal (to maximize profits) or lower than normal (to keep a third gas station from opening up). What matters is that the prices are being fixed.

If you slice and dice price fixing, you can come up with the same conclusion as you have. And yet price fixing is considered illegal and bad.

https://en.wikipedia.org/wiki/Price_fixing

In the case of LIBOR, it is a situation very similar to price fixing, but at an imaginably large scale. That is why it was a huge scandal and that is why people and institutions got punished for it.

> What is not legal is if even two or more has station owners make a secret pact to keep prices the same.

This isn't analogous, because the gas stations aren't repeatedly buying gas and selling it back to other gas stations within very short intervals. They're buying it from suppliers and selling it to consumers who use it. Banks are buying and selling liquid assets (money) to/from each other overnight. The same banks alternate back and forth between being buyers and sellers (and some may even act as both buyers and sellers within the same trading period). So "price fixing" doesn't have the same effect as it does in other markets, because it's essentially a closed system.

It's weird to talk about Libor as collusion because, again, the entire point is to save banks the trouble of having to run around and find out all of the possible rates that each bank would charge each other. Instead, they can use Libor as a estimate of the "going rate", and with the caveat that everyone knows it's not always accurate, they can figure out their own pricing accordingly.

"So "price fixing" doesn't have the same effect as it does in other markets, because it's essentially a closed system."

Except that it's not a closed system. That would be the point.

> Except that it's not a closed system. That would be the point.

Overnight interbank loans are a closed system, by definition.

The part that isn't a closed system is when people use Libor as a metric for other purposes. Which isn't a reasonable complaint, because that's explicitly not what the Libor was created for, and it was never intended to mean anything else.

That's like saying that the Internet was created to connect military computers, and therefore it's not reasonable for civilians to complain when governments spy on their internet activity.

That is: Things grow beyond "what they were created for". To say that any such use is fair game for abuse is both unreasonable and unwise.

> That's like saying that the Internet was created to connect military computers, and therefore it's not reasonable for civilians to complain when governments spy on their internet activity.

No, it's not at all.

It's more like when people complain that memcache doesn't make a good tool for session handling, or that it's not fault-tolerant and has zero replication. Of course not - it was never meant to be used that way, and the documentation explicitly says it shouldn't be used that way[0]. If you still want to, fine, but don't complain that memcache is a bad system when you're simply using it against its intended design.

[0] https://github.com/memcached/memcached/wiki/ProgrammingFAQ#h...

Your analogy says that the use was incompatible to the original intent; mine says it was outside the scope of the original intent.

So, to see which analogy fits better: Is there a document (issued by someone authoritative) that says that Libor should not be used the way it was (i.e., to set interest rates for non-interbank loans)? If not, then what we have, essentially, is your claim that it shouldn't be used that way, against everyone using it that way.

Overnight interbank loans are an important part of banks' risk management strategy, which is why they need to know them during trading hours and why they got written in to other contracts.

That's a good point, but I don't think it fully addresses what that parent was saying, which is that Libor isn't a price, but a subjective assessment. So it would be more like if two gas stations agreed to use the same marketing language to promote their stores [1] -- yes, it's collusion, but over a signal they never represented as an objective benchmark, even though people started using it as one.

[1] "we'll both say we have 'fresh produce', and one month out of the year we'll both say it's 'very fresh produce'"

Libor is in fact a price - the price of renting money. There are actual loans and/or bonds made at Libor (often plus an offset).

The point is that there is a subtlety -- the difference between a price, and a signal that other people use to set prices. Libor is the latter. Colluding on a signal is not the same as colluding on the prices themselves, the argument goes.

A lot of large loans are set to Libor plus an offset. You might think it is a signal, but is, in a very real sense, also a price.

Well... SilasX has at least half a point. If the Libor rate is being messed with (and if the messing nets to a Libor that is a consistent, say, 0.05% too low), people making contracts could adjust the offset up by 0.05%.

But I still think that SilasX's view isn't the right way to look at this. Why were people messing with Libor? It wasn't because they were going to make more money off of interbank loans. It's because they were going to make more money off of other things that were keyed to Libor. So, yes, they were using Libor to manipulate the price of things.

And if I understand correctly, it works like this: The contract says Libor + X%. The X is fixed at the time the contract is negotiated, not at the time it closes. Then, later, the contract closes, and uses the Libor at the date of closing. This means that the Libor-setters move last, after the offset is determined. This makes it impossible in practice to use the offset to fix the Libor machinations.

Agreed, in that every loan I ever saw had a fixed Libor offset. Disagree, in that Libor was fixed at closing, instead it was most often used as a floating rate. So, you had a floating Libor + a fixed offset. In regards to manipulating Libor to move downwards would have only hurt the bottom line of my firm.

Indeed. I worked at a hedge fund that also traded bank debt. One of the loans we owned while I was there was offered at Libor + ~18% (yes, percent, not basis points), and the notional was a half billion pounds. The daily accrued interest on that loan paid my yearly salary twice over. This loan was also held in the same era that the article is regarding. The US dollar vs pound exchange rate was extremely volatile in that same period. That period resulted in lots of late night phone calls regarding wild PnL swings and what happened and why. I had never really paid much attention to FX rates until that period. But, when the daily question is why are we missing X millions of dollars, and the answer was due to a fall of the dollar vs the pound, I had to start paying attention. That year or so was the worst year of my professional life.

> Since mortgages are often linked to Libor (in some countries) that means all this probably means a bunch of people saved a few pennies (literally) on their mortgage payments a few years back, mostly at the expense of large investors. Yeah, I know, that doesn't mean it's either moral or legal but...stealing an almost immeasurably small amount of money from a hedge fund and giving it to a home owner is not exactly "robbing widows and orphans" territory.

Many of those large investors are mutual funds and pension funds. Some of which do fund widows and orphans. E.g. I think CALPERS was involved in one of the lawsuits?

Stealing a penny each from a million people surely must be just as evil bad and wrong as stealing a million pennies from one person (any sensible way of relating money and harm comes up with that result). I think the reason it seems less bad is that our monkey brains aren't good at comprehending the concept of millions of people, so when we say "a million people" our moral intuition just interpreters that as "lots of people", i.e. "a couple of hundred people".

True, it was also pension funds. Okay, so...at a big picture level, a million people borrow money to buy houses, and save for their retirement via a pension fund that invests in mortgages, and then nefarious activities at the bank causes mortgage rate to shift 0.05% down for a month, causing them in aggregate to gain thousands of dollars as homeowners, but lose thousands of dollars as savers, for a net change too small to really measure?

It's not really the crime of the century, is it?

> Stealing a penny each from a million people surely must be just as evil bad and wrong as stealing a million pennies from one person

Arguable. Declining marginal value of money says that I'd be much more upset about losing the millionth penny than the first, so in some ways stealing a million pennies from one person is worse. (This is the moral justification for progressive taxation.) But okay, let's accept that for now. In which case...

...Libor manipulation still doesn't seem that bad? As you point out, there was a lot of overlap in the beneficiaries and the losers. Because even if stealing a penny each from a million people is the same as stealing a million pennies from one person, giving the pennies back again is still pretty different. :)

   Libor manipulation still doesn't seem that bad?

Actually, it does seem pretty bad. It was collusion by people who should be transparently honest, in/near the core of a system that is fundamentally based on trust. If that trust is weakened to much really, really bad things happen.

This isn't about the measurable effects of the dishonesty, it's about the integrity of the system.

> Okay, so...at a big picture level, a million people borrow money to buy houses, and save for their retirement via a pension fund that invests in mortgages, and then nefarious activities at the bank causes mortgage rate to shift 0.05% down for a month, causing them in aggregate to gain thousands of dollars as homeowners, but lose thousands of dollars as savers, for a net change too small to really measure?

Well, the bankers took their cut. If, using the numbers from the article Hayes made $70 million (and sure, some of that was probably acquired honestly, but let's assume maybe 10% came from this fraud), then on the one hand he's taken a tiny fraction of the $350 trillion he's been meddling with - most of the "collateral damage" moving one way or another balances out, sure (though at the individual level it's still unfairly damaging some people (those who don't own houses) even as it unfairly benefits others - stealing a bunch of money and giving it all away would still be a crime). On the other hand, buried in the middle of those mostly-netted-out effects he's had when you look at each one individually, he's stolen a total $7 million, which is nothing to sniff at.

I honestly don't get how you can be all smiles while apologizing for white collar crime, all the while knowing that white collar crime of this variety, or of wage-theft variety, has so many more second-order effects than "regular crime".

Maybe not you, but I've met other people who don't end up with the corollary that such things don't matter, but who instead seem to set out from the get-go to "not be one of those people hating on the 1%". It's like the hipster "I don't like music that's popular" but transposed to "I'm so learned that I actually rationalize white collar crime".

Agree. But the libor scandal is dwarfed by how much the Fed has robbed the widow and the orphan to give to the "struggling borrowers" (i.e. over-leveraged property speculators), through artifically low rates and massive money printing.

I don't disagree. But that's by and large what the public voted for, and those regulators are at least as far as we know acting in good faith, not trying to earn fees for personal gain.

It's far more complicated than that. At a first pass, the drop in interest rates helped debtors who were on variable rates or able to refinance, and creditors who held fixed-rate assets. Conversely, fixed-rate debtors and variable rate creditors got hurt.

The drop in rates arguably raised asset prices, which helps e.g. people who currently own houses. This comes at the expense of people who need to purchase houses in the future. So there are generational implications there.

However, it can also be argued that the drop in rates catalysed an economic recovery, without which there would be fewer jobs for the younger generation.

All of this is arguable. But to say that the fed is robbing widows through artificially low rates and money printing is just emphatically wrong.

>The drop in rates arguably raised asset prices, which helps e.g. people who currently own houses. This comes at the expense of people who need to purchase houses in the future.

>However, it can also be argued that the drop in rates catalysed an economic recovery, without which there would be fewer jobs for the younger generation.

trickledown houseinomics?

Stealing an almost immeasurably small amount of money from a hedge fund and giving it to a home owner is not exactly "robbing widows and orphans" territory."

The amount of money involved was hardly insignificant. Per WP:

"Early estimates are that the rate manipulation scandal cost US states, counties, and local governments at least $6 billion in fraudulent interest payments, above $4 billion that state and local governments have already had to spend to unwind their positions exposed to rate manipulation.[53] ... Timothy Lee, a capital markets expert at the Federal Housing Finance Agency Office of Inspector General, said in a 3 November memo that Fannie Mae and Freddie Mac may have lost more than $3 billion because of the manipulation.[57]"

And I hate to break this to you, but the people who benefit from these schemes weren't "giving [their gains] to home owners."

There's no way you can say the magnitude of the crime justified it, especially given the small scale and vague legal situation.

As to the crime, it's called "market manipulation", and there's nothing at all legally vague about it. These people knew what they were doing, and by and large, were perfectly aware that it was agains the law. The just didn't think they'd get caught.

They could have used eurodollars but as you note it all began long ago and nobody could be bothered to change. And these fixings were not even far from whatever rate was believed to be their true rate ((for how much and from who? Always left out). Why anybody would use it in swaps was crazy. And the fx fix outlived its usefulness in the early 80s yet some clients still insisted on using it.

The flip side of libor is the rage against spoofing. As long as the price is available to be dealt on for some minimum time, who cares? You think its wrong then go deal on it. In the days when fx and money/bond markets were not totally automated trading was all about psychology- which side would cave first. Pushing prices and then calling off the weak side was just good market making.

> Libor was "fixed" by design. Literally, the concept behind Libor was to ask some guys to make up a rate they thought they could borrow money at (note: Not the rate they WERE borrowing money at). .. >Later it began to be used for other purposes, but it seems like a lot of people forgot that it was, at base, asking some bankers to make up a number.

To be clear, it was never intended to be a measure of credit quality for banks. It was supposed to be a measure of supply and demand in the money market, a substitute for the Fed Funds rate, only for the non-US domiciled dollar market.

Yes exactly. Market participants all knew what Libor was and how it was computed. I mean some tenors never even traded - they were always made up!

"(note: Not the rate they WERE borrowing money at)"

Why didn't people and banks just use this number?

They do now, but the problem is that a bank does not necessarily raise 11 months money in every major currency every business day. So you would end up with patchy submissions. Not sure what approach they now use for that. I presume they must do some form of interpolation.

The other thing is that Libor is an ancient instrument, created before capital markets had the size and liquidity they have today. The market must have been even patchier then.