Let's talk some econ/finance ****

[VetLegion]

Speaking of econ/finance stuff, does anyone have a view on banker bonuses and what to do about them?

I'm closely following what you Brits are doing. I'm not expecting USA to do any
meaningful reforms, but Mervyn King has uttered some fighting words, and then
there's that banking commission you set up... report is pending in September I think.

My opinion is that the current system is dangerous and something should definitely
be done (I'm not yet quite sure what).

[VetLegion]

Correct me if I'm wrong, but what downside to bondholders? Even Ireland bailed
out bondholders, which is nuts if you ask me. Bank bondholders seem to have an
implicit guarantee from the governments.

[KrazyHorse]

For example, a corporation could be entirely racist and sexist and be great at making money.

A lot of people would invest in it, because they want to make money and if they don't, someone else will.

But many of those same people would vote for law makers who would make sexist and racist practices of the company illegal.

There is nothing wrong with this.

Liking how a company is run and liking how they make money are two (very) separate things.

JM

Management indulging racist behavior would depress share price, Jon (well, actually management indulging racist behavior beyond that demanded by their customers).

[KrazyHorse]

Now, let's remember a few things (from a US banking perspective):

1) The cost to taxpayers from bailing out big banks is negative
2) Even the cost from bailing out AIG is probably going to end up being nil
3) There may be some small cost to taxpayers from bailing out smaller banks
4) The biggest externality apparent to me is the cost offloaded from small banks who have gone under in record numbers to the surviving members of the FDIC, in particular the big commercial banks who are now being assessed at higher rates for no reason other than their size. Small banks have gone under, in general, because they made and held bad commercial (and to a lesser extent residential) real estate loans
5) In general, one might rightly chide me for (1) through (3) as it appears to be ex post reasoning. However, to a large extent the agencies of the US federal government are themselves responsible for the macroeconomic climate. Certainly for nominal variables, which are the most important determinants of the health of banks who have most of their liabilities denominated in nominal terms

Now, explain to me again: how do we get from these 5 points to the thought that the structure (or even more ridiculously, the level) of wall street compensation is causing some sort of massive externality?

[KrazyHorse]

BTW, I am less familiar with the situation in Ireland. It may be that there the banks are actually insolvent. It should be noted that in that case (as in Spain) the government is NOT in control of nominal variables...

[Drogue]

Do you not know what an externality is? Bondholders are bondholders by choice.

That doesn't make them not an external party to interactions between shareholders and employees. Externalities are any effects that occur to parties with no decision-making powers over a transaction, which applies to bondholders when shareholders set remuneration that increases risk.

For decision making to be efficient, the decision maker must face the full consequences, positive and negative, of their actions (or at least have their incentives aligned with these consequences). Shareholders incentives aren't, and the way they set employees incentives means they're not either. As such, bondholders (to some extent), lenders of last resort (to a much greater extent) and the general public (to the biggest extent) get screwed, as the consequences to them aren't taken into account.

It's pretty clear there's a whopping great market failure here, however what the best way of fixing it is, or even if fixes might cause other failures, is much less certain.

[KrazyHorse]

That doesn't make them not an external party to interactions between shareholders and employees. Externalities are any effects that occur to parties with no decision-making powers over a transaction, which applies to bondholders when shareholders set remuneration that increases risk.

It makes them an external party to any given transaction, a choice they made WHEN THEY LENT MONEY TO THE BANK.

It's like claiming that my insurance company is suffering externalities when I choose to drive.

[KrazyHorse]

By the way, it "being clear" to you that you've observed some kind of market failure is less than convincing, given that you've now claimed bondholders as suffering potential externalities to every single transaction every single company in the world (that has debt) engages in.

Oh, by the way are you also claiming that mortgage companies suffer externalities when the people they've lent money to choose to leave their jobs?

[Drogue]

KH makes some great points here, but I have to stress a few things:

Now, let's remember a few things (from a US banking perspective):

1) The cost to taxpayers from bailing out big banks is negative
2) Even the cost from bailing out AIG is probably going to end up being nil
3) There may be some small cost to taxpayers from bailing out smaller banks
4) The biggest externality apparent to me is the cost offloaded from small banks who have gone under in record numbers to the surviving members of the FDIC, in particular the big commercial banks who are now being assessed at higher rates for no reason other than their size. Small banks have gone under, in general, because they made and held bad commercial (and to a lesser extent residential) real estate loans
5) In general, one might rightly chide me for (1) through (3) as it appears to be ex post reasoning. However, to a large extent the agencies of the US federal government are themselves responsible for the macroeconomic climate. Certainly for nominal variables, which are the most important determinants of the health of banks who have most of their liabilities denominated in nominal terms

Now, explain to me again: how do we get from these 5 points to the thought that the structure (or even more ridiculously, the level) of wall street compensation is causing some sort of massive externality?

Was negative. When faced with a huge liquidity problem but a much smaller solvency one, a lender with almost infinite capacity can make a killing by providing liquidity at a high price. The US government did pretty well with TARP doing this. However if faced with a solvency crisis, the government will have to shoulder huge liabilities at a loss, like the Irish government has done. Depending on how the next crisis looks depends on whether the bailouts cost a fortune or make a profit. I take the view that guarding against liquidity issues isn't really necessary, you just need to make liquidity bailouts expensive, but guarding against solvency issues is essential to avoid lumping a huge cost on the taxpayer.

Also, the biggest externality isn't small banks or any other firms failing, it's the huge reduction in investment and aggregate demand. A smooth rebalancing is much more efficient than sharp shocks, and the crisis didn't cost much until banks suddenly stopped lending and the real economy got hit. The huge deficits weren't caused by stimulus spending primarily, but by a huge drop in income, spending and corporation tax receipts.

On the macro climate, I think you overestimate the impact a government can have and the foresight anyone has. In hindsight it's easy to say monetary policy was over-expansionary during the 2000s, however there were reasons to think it not so, mainly that inflation was kept it check. The reason for this, and for thinking that governments couldn't do too much about it, was China - cheap imports kept inflation low, globalisation kept it pretty stable across the world, and a huge glut of savings searching for yield kept real interest rates low, especially long term ones. That's why you get over-investment, over-borrowing and low inflation pretty much whatever the Fed or Treasury does.

Lastly on the structure of compensation, which hits at why this was a sharp upswing and crash rather than a more gradual, stable, smooth rebalancing. If you incentivise the taking of risk, you get more aggressive selling of mortgages and loans, a more fervent search for yield, less regard for any ability to pay back (the loans are being securitised, so why should the sellers care or incentivise their staff to care?), trading strategies become more risky, volatility rises, and any shocks are amplified hugely by these and the use of derivatives.

I'm an old-school economist. I tend to think that if you get the incentives for people right, you don't need to worry about micromanaging or strictly regulating the actions they can take. So while you could prevent crises like the last one occuring by mandating lower-risk actions and mitigation, I think you can do it a lot more efficiently by altering incentives and letting people then get on with their jobs, with the added bonus that you aren't just solving what happened with the last crisis.

Bonuses for Wall Street wasn't the main reason the crisis happened (I would argue), but incentives to take risk did exacerbate it and cause problems. Size doesn't matter at all to me, but making sure that compensation is aligned with risk-adjusted performance and internalises the externalities taking risk imposes is a significant step towards making banking safer and crises less likely, without much economic cost.

[Drogue]

That doesn't make them not an external party to interactions between shareholders and employees. Externalities are any effects that occur to parties with no decision-making powers over a transaction, which applies to bondholders when shareholders set remuneration that increases risk.

It makes them an external party to any given transaction, a choice they made WHEN THEY LENT MONEY TO THE BANK.

It's like claiming that my insurance company is suffering externalities when I choose to drive.

By the way, it "being clear" to you that you've observed some kind of market failure is less than convincing, given that you've now claimed bondholders as suffering potential externalities to every single transaction every single company in the world (that has debt) engages in.

Oh, by the way are you also claiming that mortgage companies suffer externalities when the people they've lent money to choose to leave their jobs?

Clearly you have no idea what an externality actually is. Yes to all, they are all externalities - the whole idea of moral hazard was created because with insurance the person making the decision doesn't face the cost of their actions, and this is why you have things like excesses and limitations of cover to stop people acting recklessly when they're insured. I mean come on KH, at least read a little about externalities, moral hazard or some of the other terms that even the press throw around about the crisis. You just look silly using technical terms incorrectly and then trying to call up others who actually do know what they mean and use them properly. You're better than this (I mean that sincerely, at least from my hazy memories of Poly of old).

The issue isn't whether there are externalities - as you say, there almost always are to everything - it's whether they're big enough to matter and thus whether it's worth going through the hassle and inefficiency of attempting to internalise them or otherwise solve them. That there are externalities here isn't the question, the question is whether they cause enough damage to warrant addressing them, given that intervention is always at best a second-best solution.

[Kuciwalker]

Drogue, you have no idea what the **** you are talking about.

In economics, an externality (or transaction spillover) is a cost or benefit, not transmitted through prices, incurred by a party who did not agree to the action causing the cost or benefit.

The probability of the action was part of the price of the bond.

[Drogue]

Drogue, you have no idea what the **** you are talking about.

In economics, an externality (or transaction spillover) is a cost or benefit, not transmitted through prices, incurred by a party who did not agree to the action causing the cost or benefit.

The probability of the action was part of the price of the bond.

Seriously, will people read before they try and act like experts. The probability of action is transmitted, but the action isn't. The buyer makes a guess at how likely the seller side is to do certain things, but the seller gets to decide, and that decision (and the millions of others that aggregate to form a risk level) is not priced in. It's a moral hazard - the seller can take action after the buyer has chosen whether or not to buy and the price has been set that affect the value of the product. Just like insurance. This is the key difference between asymmetric info (like moral hazard) and imperfect info, and why the former is a problem but the latter isn't, and I'm going to explain this in the hope that you guys stop misunderstanding how an externality works:

Let's assume you sell me a bond that pays out if some random company doesn't default. You have no control over that company, and nor do I, so we both make a bet on the probability of them defaulting and this gives us a value of the bond and thus a price. The market is efficient, there's no issue.

Now let's assume you run a company and sell me a bond that pays out if that company doesn't default. You do have some control over that (by deciding how riskily to behave), but I don't. Now I could still make a bet on the probability of your company defaulting, however you can then decide to change your risk level after I do that. Since you would get all the benefit of running more risk, yet some of the cost would be passed onto me, you have a greater incentive to do that.

If you owned the company entirely and had no other debt or equity holders, you would have efficient incentives. Once you sell bonds, and are able to change your risk level after you do, other people take some of the downside risk but get none of the upside risk from you doing so. This is clearly an inefficient incentive. Now, you could remove this issue by mandating that your risk level doesn't change once you sell bonds (they could demand this in their contract), but that would be hugely inefficient as you wouldn't be able to take advantage of opportunities or change as the market changes.

Just as with insurance, if someone can alter the risk after other people have taken on some of that risk, that is an externality and you have a moral hazard problem.


I have no issue explaining this to people or having a debate about whether in this case the issue is bad enough to warrant intervention, but don't start trying to claim that people have no idea what they're talking about when it's you who hasn't understood what it is. You have to be really ****ing sure you're right before you start telling others they have no ****ing clue. Great, you can Google a definition and use it, but don't try and act like you know so much about it when, as is pretty clear in this case, you don't fully understand what it means in practice. Yes, I realise this bit is a bit patronising, but after trying to tell me I don't know what I'm talking about when this is my day job and I know pretty damn well how it works, I think that's justified.

[Kuciwalker]

I have no issue explaining this to people or having a debate about whether in this case the issue is bad enough to warrant intervention, but don't start trying to claim that people have no idea what they're talking about when it's you who hasn't understood what it is.

Are you an idiot, Drogue? That's exactly what you did to KH.

And it's not a case of "if the issue is bad enough to warrant intervention", IT CAN'T POSSIBLY BE BECAUSE IF IT WERE THAT BAD PEOPLE WOULD JUST NOT LEND MONEY.

Moral hazard and externalities are distinct concepts: one is avoidable and one is not.

[notyoueither]

I am an industry expert, doofus

Seems quite appropriate in this thread.

[Kuciwalker]

Bonuses for Wall Street wasn't the main reason the crisis happened (I would argue), but incentives to take risk did exacerbate it and cause problems. Size doesn't matter at all to me, but making sure that compensation is aligned with risk-adjusted performance and internalises the externalities taking risk imposes is a significant step towards making banking safer and crises less likely, without much economic cost.

Or we could let the ****ing stockholders design their own incentive schemes, since they have every incentive to do so. And lenders have every incentive to price the moral hazard into their bond purchases. Given your belief in the importance of incentives, what is the case for government regulation again?