Dr Setser writes:

“He instead opted to highlight that the various forces that Nouriel and I argued might make Bretton Woods unstable are still in play.”

That’s certainly what Paul Volcker believes:

‘ “It’s incredible people have gone on so long holding dollars,” Volcker said during a panel discussion at the event. “At some point, you will get a situation where people have had enough,” he said. He added that he wasn’t ready to “extend” a previous prediction of a crisis within two and a half years.’

http://tinyurl.com/yf77zo

I agree that you’ve been wrong, Brad. There’s a reason why you guys write papers, give advice, and argue with other “smart guys” – all safely from within the comfort of the fantasy world that theory and academia provides, never having to make money in the big bad real world by betting along the lines of your assertions.

Anonymous … yeah, i argue “safely within the comfort of my fantasy world” as you say. On the other hand, you argue safely within the comfort of the hard to identify moniker anonymous. That doesn’t really impress me. Pick a handle.

I put my intellectual reputation on the line pretty much every day — and try to be upfront when i get things wrong. Some folks care a lot about money, i care a lot about ideas. So if I get something wrong, it does hurt — even when i don’t have money on the line (and sometimes i do, tho the amounts are small — like lots of others, I have lost money with RMB NDFs).

i have been upfront that I got two big things wrong back in early 2005. I thought rapid Chinese reserve accumulation would bleed over into higher inflation, and a set of domestic problems would develop to weaken China’s commitment to the peg. that didn’t happen in the past two years. i have never shied away from admitting that, and have tried to explore why I got that part wrong. Folks in the markets occasionally make bad bets too — they key is to make more good bets than bad ones.

The other thing i got wrong was the GCC’s willingness to build up its dollar portfolio. I expected the same political factors that make the GCC reluctant to hold $ in the US financial system would make them reluctant to hold dollars generally. It seems like i got that wrong too.

So, if I may, a question for you and truthhurz — a lot of the things I thought would appear in China now seem to be materializing in some GCC countries. Qatar and the UAE specifically. Rapid credit growth. Strong money growth. And rapidly rising inflation (even with price caps on gas). They are doing less fiscal sterilization, but don’t seem to have stepped up monetary sterilization — and doing so would push up local interest rates, attracting more capital inflows and the like. Do you all have a view on whether this is sustainable? And if so, for how long? Will the GCC opt to slow down spending to avoid a real appreciation with more fiscal sterilization? Will it take other steps to cool the economy? Will it allow real appreciation thorugh high inflation? Or will it instead opt for some form of revaluation against the dollar?

All are difficult questions. What is your bet?

And do you all have an answer to the “one trillion dollar question” — will China be willing to add another $ 1 trillion to its reserves over the next four years (maybe three) to sustain the current global equilibrium, and continue to hold roughly 70% of them in dollars? What is your bet on that?

US consumers will soon exhaust their ability to service more debt. Its like watching a drunk with an ample supply of liquor and trying to predict when he’s going to fall down. Regardless of when he falls, he’s going to have a hell of hangover.

RN, It’s pretty obvious the US is heading toward deflation, and will most likely bring down the global economy with it. I don’t think the GCC or China need to do a thing – money growth will be cooled by the US bust. Once contagion from the US banking sector spreads globally, the credit markets will piss on the laughable deals being underwritten around the world. US Govt debt should do just fine. The chain of events works, very simplistically, like this: No more fools interested liar loans in the US, no more growth in the gulf states or in the east. But the price of oil now is telling you this already.

TruthHurtz,
Care to tell us why deflation is the obvious destiny? China for one, has plenty of stimulus ammunition in reserve. If China loses its external growth engine, the party will have no choice but to switch to domestic stimulation for survival, and China could easily export inflation next. It already did for the commodities. And China’s obvious big stimulus is car ownership, which will drive oil consumption. There is a reason why it has been so desperate in locking down oil supplies (it certainly doesn’t need it for energy — it has plenty of coal).

Brad–Since I basically agree with you on the unsustainability of the Bretton Woods 2, I just would like to encourage you to sharpen your analysis.

Since Chinese accumulation of dollar obligations appears to go on and without much change, I wonder if we might not have to wait for some external crisis to bring about that change or breaking point. What might happen if there were a military crisis in the Middle East that sent oil to over $100 a barrel and kept it there for some time?

RN:
thanks for the scum comments, takes one to know one. how’s that for sophistication?

secondly, I love brad and brad’s blog. he is the best thing going on the web. yeah, i can be rough with language. i know my faults. i’m a devil. but i do love this blog in way you can’t understand. i love this it in the same way that ike loved tina. and hey, mtc, i even love you too. it’s such a shame that because of your pediphilic tendencies, your love goes forever unrequited.

It’s time to go back to basics:

http://en.wikipedia.org/wiki/Inconsistent_trinity

“Inconsistent Trinity (or Impossible Trinity) is a term used in the macroeconomic policies for an open economy. It represents the impossibility for a country simultaneously to maintain a fixed exchange rate, to permit free capital flows, and to have a monetary policy directed toward domestic objectives.”

http://en.wikipedia.org/wiki/Robert_Mundell

Quick question: is China lending funds twice, internally and externally? Are USD reserves also collateral for internal loans.

“tc, i even love you too. it’s such a shame that because of your pediphilic tendencies, your love goes forever unrequited.”

My God, you really are a complete low-life.

Statistical accounting manipulation of US Economic Statistics by Bush Administration

“Today’s investors face a difficult choice, one they’re not to be envied for. They can see the relative weakness of the US economy and they’re registering the tectonic shifts in the world economy. They know that a great statistical effort is being made to prolong the American dream. For some time now, government statistics have announced sensational productivity leaps for the US economy — productivity leaps that, strange as it may seem, haven’t led to any rise in wages for years. This is in fact genuinely bizarre: Either capitalists are reaping the fruits of increased productivity all by themselves — which would be a political scandal even in capitalism’s heartland — or the productivity leaps exist only on paper. There is much to suggest that the second hypothesis is correct.

Half the world is impressed by the low levels of unemployment in the United States. The other half knows that these statistics aren’t official, but the result of a voluntary telephone survey. Many of those who declare themselves employed are assistants and day workers. Working just one hour a week is enough for one to be classified as “employed.” Given that it’s considered antisocial to declare yourself unemployed, the US statistics may well say more about American society’s dominant norms than about its actual condition.

The US economy’s high growth rates aren’t to be completely trusted either. They are the result of high public and private debt. In no way do they express an increased output of domestically produced goods and services that the United States has achieved by its own strength. They say more about the successful sales ventures of Asians and Europeans. New loans taken by the US government were responsible for fully one-third of US economic growth in 2001. In 2003 they were responsible for a quarter. The United States is an economic giant on steroids — doped so its decline in performance doesn’t become too apparent.”

– Gabor Steingart, America and the Dollar Illusion, October 25, 2006
http://www.spiegel.de/international/0,1518,440054,00.html

Stormy,
Maybe you forgot that Chinese consumption is growing much faster than U.S. Chinese structural problem has been discussed here before. The problem is that the Government controls even more savings than the high-saving households. But it is much easier to distribute wealth than to extract wealth, to lower tax than to raise tax. There is no reason why this structural problem can not be corrected in a hurry.

Brad-
You should charge $20,000 for your posts.

I love this site – not only for the wisdom of Brad but also for the wisdom and diversity of comments.

Speaking of diversity, I am among the 2% of commentators who believe that the trade deficit will be reduced by direct action of the U.S. government to limit imports – market forces will not act soon enough to satisfy the public.

Not an accepted view today. The 2006 election showed that voters can take an abrupt turn when they become convinced that current practice is wrong.

JW — I like your thesis that three years from now BW 2 will be gone, but no one will be sure when it ended. tis an interesting hypothesis.

that said, i think the peg/ no peg issue is a red herring. China still intervenes massively to keep the RMB from appreciating. It now looks like a crawling peg with a small band. No matter. What counts is the scale of intervention.

finally — I don’t charge 20K for a post, but i do charge for scenarios about the future (seriously) … I actually have done forecasts looking forward on the next trillion and the like. But, alas, I can not give them out for free and do this blog as well. Too bad — putting material in the open and getting reactions is better intellectually. tis less lucrative tho.

” For more than five years the American consumer has been the “cash-cow” of US growth, contributing more than 70% to the resulting progression of the United States economy. Stimulated by the easy money policy promoted by the US Federal Reserve in order to avoid a catastrophic recession feared after the explosion of “internet bubble”, the US consumer rushed into a frenzy of purchases of consumer goods. Encouraged by banks and the whole US financial system, he exceeded his own financing capacities and plunged since 2004 into generalized debt [7] and into a situation not seen in the United States since the dip of the Great Depression post 1929, namely a negative saving rate [8].

The Federal Reserve, the Bush administration and the republican Congress, as well all the financial and banking sectors of the country then fed the fiction of a continuous and fast enrichment via the development of the “real estate bubble” which convinced the majority of the country’s middle class, including its least ‘well off’, to rush into a strategy of purchase, and often of speculation, in real estate [9]. In parallel, the very strong growth of real-estate prices made it possible for the financial sector to offer loans fo

“3) I think what has made this all so hard to predict using economics is that the behavior of the Chinese government is constrained by political factors. I notice that the sensible suggestions I read in many economic blogs about what China should do never come up when I read experts on Chinese politics.”

Kevin:

One very possible explanation is that those “experts” on Chinese politics don’t know what they’re talking about.
Also, one can also say almost the exact same thing about the need for US to do sensible things but they never come up when reading experts on US politics.
I suspect that things are very complicated and the other side on this debate is putting up very good arguments. Have you read McKinnon or Stiglitz lately?

TruthHurtz (and Anonymous),

Speaking as someone else who finds bsetser and his blog very enlightening, I think your substantive points would be better taken if your behavior were slightly less boorish. My hastily written posts have gone over that line, I think, a couple of times, and I have always regretted it. You may disagree with bsetser’s entire perspective, as in many ways I do, but the essence of civilization is the ability to talk civilly with people you disagree with.

If you compare the comments on the Setser blog to the Roubini blog, for example, you’ll note that the signal-to-noise ratio is much better here. This is probably just because Roubini is the “headliner.” But if you pollute this public space by crapping on the rug and sticking gum to the furniture, all the intelligence you value here will vanish. Never mind that you’ll discredit the exact point of view that you support.

You sound as if you have interesting points to make. At least you sound that way to me. I wish you would make them, and skip the alt.flame rhetoric. It is not new – to paraphrase the Beastie Boys, I was flaming people when you were s***ing your mother’s d*** – it is not interesting, and it drives interesting people away. Please desist.

JW,

Taleb is a god. A golden god.

The reason people worry about hedge funds and derivatives is that via derivatives, hedge funds have found ways to capture reward and socialize risk. As a result, the state is left insuring an increasingly large pool of risk. Or, if you prefer, it is increasingly a seller of volatility.

If it protects this risk, its actions are dilutive, aka inflationary, and encourage the exact behavior it seeks to discourage.

If it disowns the risk, it instantly devalues an enormous set of financial instruments, causing a panic (aka debt deflation). Moreover, it cannot convincingly disown the risk without actually permitting a panic. Its most recent chance to do this, for example, was the LTCM crisis. Perhaps prudently, it balked.

This is political zugzwang. The only way to delay the inevitable collapse of the system is to make it larger and more complex – for example, to add more regulations on hedge funds and derivatives. At a certain point this becomes impossible (see under: Basel 2).

This is a point at which the discontinuity, Taleb’s black swan, may appear.

More generally, the discontinuity happens when, and only when, everyone expects it to happen. There is no good way to predict the evolution of this chaotic feedback loop. However, if you are looking for a proxy, probably the best one is the percentage of the world’s money that has been exchanged for assets that are likely to retain their value in the event of a discontinuity.

It strikes me as very unlikely that this indicator can get anywhere near 100. Frankly, I’d be surprised if it reached 10.

As I’ve said previously, it is essential that someone construct and execute a “flag day” plan for returning to a stable financial system, probably based on commodity currency, before this event. Otherwise I think the prospects of our present political system are rather dim. I am certainly not in love with said system, but one could definitely do a lot worse.

Just an additional comment:

I think that the “calls to buy oil, gold, silver, and other rare metals” in China are all hot air. The PBoC is not a profit-making institution. Its goals are exclusively political. If it wanted the dollar to collapse, it would have cut the rope already. I can’t think of any policy action more counter to the philosophy of today’s Chinese government than intentionally committing such a destabilizing act.

Despite these little strains, all the world’s central banks are essentially allied in their effort to preserve BW2. It is in all their interests and they all know it. By threatening to buy precious metals, the Chinese are just engaging in a little CB-to-CB jawboning. It’s not unlike threatening to invade Taiwan – it makes by far the most sense if you think of it as deliberate communication, rather than any kind of evidence of actual intention.

The source of all destabilizing pressures is the private sector. The golden goose has turned into a golden eagle. It still lays its famous eggs, and they are only getting bigger. The CBs eat egg for breakfast, lunch, and dinner. They are so done with egg. But where does it stop? Aquila chyrsaetos is by no means the world’s largest conceivable bird…

bsetser: I do have a somewhat different take on things. The difference for a bank between 0.0% volatility and 0.01% is huge, and I think the Chinese government actually moving as quickly as it can to appreciate the RMB without cause a major financial crisis. Basically, once you unpeg the currency, you need a huge number of systems in place to manage currency fluctuations. Business and banks have to develop new systems in order to deal with the new world, and this takes time. (You got to buy the computers, test the software, train the traders, hire the accountants, deal with the disasters that happen when people play with power tools for the first time etc. etc. etc.)

Pulling away from the peg gradually lets these systems have time to develop and mature, and my grand prediction (tada) is that over the next two years, you’ll see much more in the way of fluctuation in the RMB, and gradually you’ll reach a point in which the RMB is purely floating.

moldbug: I don’t think that derivatives are transferring risk to society, I think that the direction is in the other way. Let me give you an example.

In China, you can’t get a fixed interest mortgage. In the United States, almost everyone with a long term mortgage has a fixed interest mortgage, and even people with ARM’s have caps. This means that if in 2012, the government has to raise interest rates to deal with the aftermath of the Icelandic crisis, some rich guy will lose a small fraction of his wealth, and I won’t lose my house. Banks would not be able to offer fixed interest mortgages without interest rate derivatives.

Without a market in derivatives then it would be impossible for the government to adjust interest rates much without fear of cause a whole bunch of defaults in homeowners, and we return back to the 1960′s when interest rates were effectively fixed by the government. The problem with that system is that it blew up in the 1970′s when it just wasn’t flexible enough to deal with the oil shocks.

Also, I think you are shooting the messenger. Reality is inherently complex, unpredictable, and random, and any financial system that reflects reality is also going to have those characteristics. Imposing simplicity on a complex system is something I’ve found to be generally disastrous. A cremated corpse is easy to analyze and understand. A living human being (or economic system) is massively complex, poorly understood, and capable of surprises (both good and bad).

One other thing, it about a generation (1965-1985) for the system of floating interest rates and floating currency exchange to evolve and about a decade for BWI to collapse. I think people are being just a little impatient at trying to get China to compress that evolution into one week. The other interesting thing is that the financial system basically evolved without anyone really designing it.

There isn’t a single economist in 1965 who would have been able to predict what the financial system would have looked like in 1985, and I think that there are going to be a lot of surprises in the post-BWII era.

JW,

A free market in derivatives would be just fine, but I don’t think that’s what we have.

Every derivative has a counterparty. In a free market, the price of a derivative must include the risk that the counterparty will default.

Let’s say I buy a derivative from Goldman Sachs. If GS defaults during 2007 or 2008, the derivative is worthless. (I am being very simplistic here, obviously.) What is the chance that Goldman Sachs will default in 2007 or 2008?

Let’s make this a variable. We can call it $gsd_0708. This is a real number between 0 and 1. Because quantum fluctuations can cause arbitrary events, the range is open – it cannot be either 0 or 1. It’s all very precise and scientific.

Now how do we calculate $gsd_0708?

Answer: we can’t. There is no conceivable quantitative procedure that will compose a value for this variable which isn’t slightly brown and stinky as though it was pulled out of you know what. It is a fudge factor pure and simple. Any calculation which depends on it is fudged by definition. Any balance sheet which uses this calculation to mark to market isn’t worth the pixels it’s printed on.

Feynman’s essay on cargo-cult science is required reading for everyone. This isn’t cargo-cult science – it’s cargo-cult finance.

Now wait a minute, you say. I know the guys over at GS. They’re good guys. Super smart. How could they default? Can’t we just set $gsd_0708 to 0?

Sure. And all this does is allow GS to sell (regulators permitting) all the derivatives it wants. After all, if you can assume $gsd_0708 is 0, so can they – both entities existing, after all, in the same reality. So GS can insure anyone against anything. It can protect you against the possibility that $rate is too high, too low, too even, or too odd. Woo hoo! Fixed-rate mortgages for everyone!

Now you can’t throw a stone without hitting some titan of economics who is warning of imminent disaster. And somehow, volatility is at historical lows. Why is this, exactly?

It’s because GS, and all the other GSes in the world, can sell all the volatility they want. It’s a race to the bottom. Why is volatility cheap? Why is anything cheap? It’s supply and demand. There may be a lot of demand for volatility, but the supply seems to be pretty much infinite. Because $gsd_0708 is 0, and it ain’t exactly in the numerator. That’s math for you.

Madness. This is obvious madness. So why do all these smart people participate in it? Are they stupid?

No. They are crazy like the fox. They have made a very simple observation: if everyone assumes that GS can’t default, GS cannot be allowed to default. That is, to eject the egregious passive voice, Uncle Sam will not allow GS to default. And since all these instruments are denominated in dollars, and Uncle Sam can print or lend as many dollars as it wants, it’s safe to say it will succeed. Thus, in reality, $gsd_0708 – in dollars – is 0.

It should be obvious how underestimating risk implies monetary dilution (aka “inflation”). You are creating an unlimited stream of liquid instruments whose market price is nonzero. You are certainly not creating any actual goods.

People sometimes complain about the US flood insurance program. How it encourages people to rebuild in vulnerable areas. They should complain about the US risk insurance program, which is building billion-dollar mansions on inflatable pool rafts tied together with Twisties and anchored with a brick and a piece of fishing line in the open ocean off Bermuda. Volatility? What’s that? Haven’t seen any volatility around here.

The only barrier to this hemorrhage is regulation. But because regulators have no precise way to define how much risk is being underpriced – which would require them to, essentially, set prices (perhaps we could have an Office of Risk Administration) – it is entirely unfeasible.

Someone mentioned impossible trinities earlier. Here is an impossible trinity for you: free capital markets, derivatives, and fiat currency.

In other words, what we’ve done is set $gsd_0708 equal to $usd_0708. Uncle Sam has effectively made Goldman Sachs part of the Treasury and let it take a ride on the big printer. As long as you assume the risk of a dollar collapse is zero, any number of derivatives is no problem.

This is also why buying a mold future backed by GS, or NYMEX, or whoever, is not the same thing as buying actual mold in an actual vault in Zurich. What’s the difference between these? It’s a small matter of $usd_0708. Which some people probably think they’re betting against when they buy the future. Whoops.

Let’s take a quick look at one way – just one way, there are lots – that Stein’s Law could operate on a system like this.

All this derivative printing means a lot of liquidity is sloshing around. I mean, we’ve basically piped the dollar’s carotids straight into the Hudson, here. This is a well-known cause of aggregate consumer price appreciation. Aka, in Fed-speak, “inflation.” Let’s say appreciation outraces the BLS’s ability to define it away, and Uncle Sam is forced to respond. Maybe they shoot Bernanke and put Volcker back in charge, and we get a rerun of 1981.

I’m embarrassed to admit it in these circles, but because I took out a mortgage back when I thought Ludwig von Mises was one of your more minor Nazi generals, like Manstein, maybe, or Model, well, what I’m trying to say is, I have an ARM.

But my ARM is capped at 10%. Which means that if rates go up to a Volckerian 20%, I may lose my shirt, but I won’t lose my house.

But what of the poor schmo who bought (never mind whoever bought all those actual fixed-rate mortgages!) my mortgage? What happens to him? Well, probably, nothing. Because he probably bought a derivative to protect him against just that.

Which was probably issued by someone just like Goldman Sachs. Who has probably sold quite a few of these instruments. And will therefore be, well, pretty much gasping for dollars. Which Uncle Sam will, because the consequences of not doing so would be even worse, provide.

But printing lots of new dollars has a well-known side effect. Which is an increase in interest rates. Which puts anyone who’s sold rate insurance under even more pressure. And so on. When Cassandra says, “inflation is the endgame,” perhaps this is what she means.

Past hyperinflations have taken months or even years. Quantitative finance being what it is, this scenario seems more likely to play out in weeks or days. It may even happen in the course of a single trading day. The open-market desk would just keep dumping money into the system, and it would just keep vanishing. Division by zero. Segmentation fault – core dumped.

There is no way to predict when this will happen. Or even whether it will happen. But there are plenty of other feedback loops that could play out similarly. Stein’s Law will prevail.

And there’s your post-BWII era. If anyone feels I’m being too alarmist, please feel free to say so.

JW,

I agree – probabilities (except in the absolute quantum-fabric sense that I was describing, which is not really relevant in the human world) are meaningless. And yes, the market can give you a number for most any kind of probability.

The question is: how do you know the numbers are right? And what do they actually mean?

First, to answer your point about bond defaults, when you make a trade with GS, the number you want is not actually the probability that GS corporate bonds will default. The number you want is the probability that someone will pick up GS’s end of your trade.

In normal circumstances these should be the same. But GS defaulting is obviously not a case of normal circumstances. In reality, I suspect there are many scenarios in which GS’s customers would come out better than its bondholders.

But this is a minor point, and you’re probably better equipped to evaluate it than me. Let me get back to the main thread.

I was being too extreme and simplistic in my example. The thumb on the scale is relative, not absolute. You are right, no one assumes that $gsd_0708 is 0. But the thumb remains.

The thumb is the moral hazard of US bailouts. This affects the market price of all kinds of risk. If you assume that no bailout will happen, the market price is wrong. Extremely wrong. (Fortunately, this is a false assumption.)

The credit spread between GS bonds and T-bills is not $gsd_0708. It is ($gsd_0708 – $usd_0708) – the probability that GS will fail, and that the US will refuse to bail it out.

As I said before, it is $usd_0708 – essentially, the probability of a dollar collapse – that matters. You may be patriotic, but your money isn’t. To an independent investor, the US is just another big company. And no financial market priced in its notes can produce $usd_0708.

Exchange rates between the dollar and other fiat currencies could do so in theory. Some people think they do so. But, of course, forex rates are heavily managed for political reasons – they tell us more about the wisdom of central banks than the wisdom of crowds. The world is quickly becoming a single fiat currency bloc.

(I hesitate to use the phrase “market fundamentalism”, but anyone who trusts these kinds of numbers is certainly asking for it. Markets are like science – they work not because they have some magical inherent infallibility, but for specific logical reasons. If these reasons are inoperative, you have a cargo cult market.)

The same is true, unfortunately, as you’ll find out if you look into the matter, of the various rates that various markets construct between dollars and mold. These numbers are not uninteresting, but I think to call them meaningful would be to go too far.

If you disagree, perhaps you should read this IMF issues paper. If the CBs would tell us exactly how much mold they still have in their vaults, as Portugal did (give it up for Portugal!), life would be a lot more interesting. Which is, I suspect, exactly why they don’t.

Other commodity prices may provide interesting bits of information. But most other commodities are not suitable, for various physical reasons, as long-term stores of wealth. When someone buys a dollar-denominated oil contract, for example, she is not exactly expressing her opinion that the dollar is likely to go poof – because her contract may well go poof along with it. If she had a big storage tank in her backyard and actually put the oil there, it might be different, but this tends to be impractical, unless your name is, say, “China.”

So besides the mournful jeremiads of Paul Volcker, Robert Rubin, etc, I don’t see any good way of estimating $usd_0708.

But wait. How did we get from derivative risk to the fate of the dollar? These are both valid RGE subjects, surely, but haven’t we digressed a little?

Not at all. Back to derivatives.

JP Morgan wrote, I believe, something like 5 trillion dollars of nominal derivatives last quarter. Bringing it to a grand total of like 50 or something.

Obviously the current market value of these instruments is much lower. Obviously many of them are offsetting. But I still contend that it defies common sense to suppose that, if the US could credibly eschew any kind of market intervention, JP Morgan would exist ten minutes later.

Why? Well, let’s start by assuming, ceteris paribus, that the US could and did credibly eschew any kind of market intervention. Excepting official programs like FDIC, SIPC, etc. The printing press will print, let’s say, M2, and nothing more. There will be no “voluntary” LTCM type affairs. In a pinch it’s every man for himself, sauve qui peut, head for the hills.

JPM does not have 50 trillion dollars. Its ability to fulfill its obligations is dependent, I’m sure, on a tremendous number of counterparties. I’m sure it can survive the default of any one or two. I am sure there is no single risk that it can’t survive.

Risk management people certainly ask: what happens if GS defaults? What happens if JPM defaults? What happens if X defaults or Y defaults?

But I think this, with all due respect, is a piece of theater that’s played for the regulators. Its purpose in reality is not to actually manage counterparty risk. Its purpose is to transfer counterparty risk to the US government. The point is to show that you are a reasonable citizen who is not taking on any more risk than your competitors. And if something goes wrong and someone has to be sacrificed to the copybook gods, it shouldn’t be you.

I’d argue that, in this non-bailout reality that I’ve constructed, the risk is not that GS will default. Not that anyone else will default. But that everyone else will default. Not that X will happen or Y will happen, but that all of A through Z will happen.

There is no reason to suppose that these risks are independent probabilities. The probability of both GS and JPM defaulting cannot be obtained by multiplying the probability of GS defaulting by the probability of JPM defaulting. Common sense, not math, tells us it’s more likely that the whole shebang will go at once. And math has no opinion on the matter.

Let’s terminate the ceteris paribus experiment and return to reality.

What is the connection between cascading derivative defaults and a dollar collapse? Why am I using $usd_0708 for both of these seemingly separate risks?

It’s because I don’t think they are separate. Because all these derivatives are dollar-denominated, any kind of default can be staved off by printing more dollars. But this both creates a moral hazard that encourages the creation of more underpriced and hence dilutive (“inflationary”) risk, and imbalances markets in ways that tend to cause more derivatives to default. This is, in my opinion, two sides of a single death spiral. If derivatives go the dollar goes, if the dollar goes derivatives go. Again, this is just my own personal opinion.

In practice, the purpose of risk management is to make sure your own risk is safely within this death spiral. The goal of GS’s risk managers is not to worry about $usd_0708. It is to make damned sure there is no localized failure that causes GS to fail. Systemic failures are not in their pay grade. GS could simply not exist in anything like its present form if it worried about a systemic event. It would not be even remotely competitive with anyone who didn’t.

And there is the rub. That is how moral hazard works. You either sign on the dotted line or you don’t, and if you don’t you need to stay away from those who do. Because they have the ring and you don’t, and if you fight them you will lose.

The financial world is dividing. It is creating two kinds of people, green and yellow. Green people are people who treat $usd_0708 as 0. Yellow people are people who keep their money in big bars of mold in Zurich. I think the historical pattern is that this division is not stable, but I certainly cannot claim to predict the date or details of its resolution.

dealers trade on a day to day basis, while pundits predict on a scale of this year / next year/ some year soon. the future holds a range of possibilities, the exact outcome and timing of which is not predictable. the pundit who pinpoints the top of the market is merely lucky. just to be spot on does not prove foresight. a blind man backing horses with a pin can win – but that does not make his predictions ‘right.’
better to make actual trades on a short term basis – and philosophise about the future on a longer term basis – without naming the year or the day of the foreseen discontinuity.
in general, we all can see this : that exponential growth within a finite system, ends abruptly. how long does the pond lily that doubles in size every day, take to fill the second half of the pond ? answer : one day. the answer is counter-intuitive and yet undeniable.
markets are unpredictable because prediction is either ignored, or it feeds back into price changes and the market adjusts.
the kennedy rule (1929) is still the best one : if the shoeshine boys are tipping shares – the show must be nearly over. the shoe shine boys (2000s) know that property prices will go up for ever because the dollar is bound to fall, so that show may be over, too ?

1929 was an intermission. The show didn’t end, it just added, and keeps adding multiple stages. There will be more intermissions and the scripts, sets and actors will keep changing. But if it ends, we’re all dead.

Well – I do not think you were that wrong Brad……….
“Simon Derrick … rather graciously decided not to dwell on the fact that my timing was off. ”

It simply depends on what should be the investment advice followed from your analysis. Your analysis and all the potential sources of instability you & Roubini identified early 2005 were clear argumentation to invest in gold which has given a 50% return since early 2005. The reason for this beeing exactly the instability pointed out in your analysis – and the instability clearly continues over the next 1-2 years (as a minimum – more likely 5 years) driving the price of gold up further. If gold can appreciate 50% the last 20 months without the Bretton Woods 2 falling apart – imaging how gold will behave when Bretton Woods 2 actually starts falling apart ….

> In other words, what we’ve done is set $gsd_0708 equal to $usd_0708.

No we haven’t. GS debt sells as a few points higher than Uncle Sam debt. There are a whole host of derivative securities that people will sell you that will make you rich if GS defaults.

The closest parallel to a collapse of GS is the collapse of Barings Bank, and the British government didn’t lift a finger to prevent that. There’s no particular reason that the Feds would bail out GS if it got into trouble. Frankly, I would think that the US government is run by a bunch of idiots if they bailed GS out.

There is an basic exchange “regulation versus autonomy.” Hedge funds are largely unregulated, but in exchange, they fail (which they do all the time), the government doesn’t bails them out, and the rules are designed so that widows and orphans are unaffected.

> Now you can’t throw a stone without hitting some titan of economics who is warning
> of imminent disaster. And somehow, volatility is at historical lows.

Because volatility is uncorrelated to the pricing of default. The market could think that GS could collapse tomorrow. This wouldn’t generate any volatility. The other thing that you need to consider is that experts could be wrong. Risk works both ways. There is the risk of something unexpectedly good happening. If you leverage yourself betting that GS will collapse, and it doesn’t, you are in a lot of trouble.

I don’t think that you are alarmist for saying something bad *could* happen. What concerns me is that you are saying something bad *will* happen. If you believe that then you are missing the point of Taleb’s book.

> In practice, the purpose of risk management is to make sure your own risk is safely > within this death spiral. The goal of GS’s risk managers is not to worry about
> $usd_0708. It is to make damned sure there is no localized failure that causes GS to > fail. Systemic failures are not in their pay grade. GS could simply not exist in
> anything like its present form if it worried about a systemic event. It would not be
> even remotely competitive with anyone who didn’t.

You are wrong about this. The goal of the risk managers of any investment bank is *PRECISELY* to worry about systemic failures, and to make sure that GS survives even if the the US government or UK government or Chinese government does something odd. The reason for this is that systemic failures can kill a bank just as easily as local failures. And GS (which is a particularly bad example for you to give since they have a reputation as one of the more level headed banks) is competively because it survives when everyone else blows up. There is a whole bunch of internal controls and cultural devices to make sure that the bank survives, and it’s building up these sorts of internal controls from scratch that is the challenge of Chinese banking.

> It’s because I don’t think they are separate. Because all these derivatives are
> dollar-denominated

No they aren’t. If you don’t trust the US dollar, then you start moving your currency into Yen or RMB or something else, and your write your deriviative contracts in something other than dollars.

> JP Morgan wrote, I believe, something like 5 trillion dollars of nominal derivatives
> last quarter. Bringing it to a grand total of like 50 or something.

Nominal derivative values are meaningless and have nothing to do with value at risk. I bet $1 that the price of gold will rise to $1000. The nominal value of the derivative is $1000, but that’s bogus since the value that I stand to lose is $1.

> But I think this, with all due respect, is a piece of theater that’s played for the
> regulators. Its purpose in reality is not to actually manage counterparty risk. Its
> purpose is to transfer counterparty risk to the US government.

I know people in risk management. They are serious as hell since they know plenty well that the bank could go up in smoke if they don’t do their jobs. Everyone in the industry knows the examples of Nick Leeson, Metalschaft, Orange County, China Aviation Oil, LTCM, and the hundreds of hedge funds that have gone bust over the years.

> But I still contend that it defies common sense to suppose that, if the US could
> credibly eschew any kind of market intervention, JP Morgan would exist ten minutes
> later.

I thought you were an Austrian.

Really interesting discussion – and it’s great to hear so many glowing references to the quality of information and fine people who are the stewards of the world’s financial system. No denying that a great deal of higher intellect goes into designing and making things work as well as they do. But:

re: “rules are designed so that widows and orphans are unaffected.”
why is the wealth gap increasing?

re: “If you don’t trust the US dollar”
isn’t this type of spin part of the theater? Is selling, let’s say loonies and buying USD, an indication that I “don’t trust the loonie”, or an indication that I think it has had a very good run and is due for a correction against the USD for many reasons.

re: “Fed and Treasury have a currently active policy of intervening in financial crises”
but what makes it so interesting, is that it is selective – it could be argued rightly so, as each situation is ‘special’.

re: “preventing that storm. Probably a smaller, but perhaps still nontrivial, portion of braintime is aimed at surviving such a storm”
the weather scenario is good, because storms are inevitable, so it is a matter of (accurately) predicting and surviving each, depending on the intensity, duration and perceived vs. real damage.

re: “But I think that people at GS, especially the risk management group, probably spend very little time thinking about how they can profit from such a storm. And I suspect they spend pretty much no time at all trying to come up with clever ways to instigate such an event.”
isn’t profiting from the storms, and perceived risk of storms, central to the game. why is the insurance industry making so much money? which leads to a strong instigation incentive, at the very least to keep everyone scared to death about the probability of any number of risks, and paying to insure against the risks – if only by talking about a bit of good theatre every once and a while?

re: “I should not have used the word “theater” to describe this, because I am sure it is done with 100% sincerity”
thought you were absolutely right to use the word theater – although I’m not sure what this would have to do with “sincerity”

JW and Guest,

Thanks – I hope others are learning as much as I am. I find it really fun to bash antlers with people whose perspectives are very different from mine. And hopefully it’s fun for spectators as well. Think of it as like one of those mantis-versus-mouse clips you get on YouTube.

JW, I have a question for you. Please note that this is a question of the actual informational variety. Ie, not the rhetorical sort – though it may help you grasp the scale of my ignorance.

In the CRM Policy group executive summary I linked to earlier, it cites one symptom of a financial perfect storm as the blurring of “the analytical distinction between market risk and credit risk.”

Now, call me crazy. But I had been cheerfully proceeding along in my life under the obviously mistaken assumption that there is one thing and one thing only, and that is risk. I had not been in any sense aware that any such analytical distinction could be drawn. And it disturbs me that people seem to be doing so, because it means either I am confused, or they are. Perhaps you can help me sort out this little conundrum.

I mean, suppose I am doing up a balance sheet and trying to assign values – numerical prices, in whatever currency or commodity – to my assets. I don’t want to sell these assets, because if I did I would have already sold them. I just want to know what a rational buyer would pay for them if I did sell them now.

If the asset is liquid, I can just use its market price (“mark to market”). I understand that in many situations accounting law requires M2M for reporting. But that doesn’t mean I can’t keep a separate balance sheet simply for my own decision-making purposes.

For example, suppose my investment strategy assigns X, some future event, a probability of 80%. But if you look at the market price for asset A, it is discounting X at 35%. It makes no sense that I should make decisions about A in a way that assumes X at 35%. I cannot simultaneously believe that X is at 80% and 35%. This is irrationality defined. If I have my own estimate, I should use it. The fact that my current estimate and the market’s current estimate differ is certainly important, but it does not mean that the two numbers cannot be calculated and considered separately.

(Remember, I could always just decide to trust the market for its estimate of the probability of X. Which I as a sensible person who digs that whole count-the-jellybeans game will probably do for A-W, Y, and Z. But if my strategy assumes I have more information than the market about X, it makes no sense for any purpose of my own calculation to accept the market’s information and ignore my own. If markets could be infallible, they would not need to exist.)

So there is no analytical distinction between liquid and illiquid assets at this level. You value an asset based on your own opinion of its present value, which can be defined as the price the market would assign it if every actor in the world had all the same information you do, and agreed with you about all of it. If your only source of information is the current price, this calculation becomes trivial. But it need not be.

Note also that the valuation criterion I have chosen, price if the asset were sold today on a market where all actors share my presumably correct views, is by no means the only interesting function from asset to price we can construct. It is just one interesting function.

For example, I can project the same function out into the future, and compare it with my estimate of what the market’s estimate will be in future. Obviously, when I do this, I am predicting future events, extending my set of hypotheticals (which is already non-empty, because my calculation already depends on this counterfactual all-clones-of-me market I am using as my criterion).

Any time you have hypotheticals you have risk, and expected value does not capture a risk profile. Even if you are Goldman Sachs, your utility function of monetary gain or loss is nonlinear. If GS was offered a bet with a 10% chance of losing $100 billion and a 90% chance of making $15 billion, would it accept?

But let’s step away from risk for a moment. We’ll be back in a sec.

Suppose asset A is someone else’s liability. It is an obligation of party Q. Suppose the risk-free value of this obligation, its expected return if Q does not default, is obvious: call it $va. So X, our critical piece of information, is the probability that party Q will default. Call it $pdq.

Therefore, I would expect that any balance sheet that one was to construct, if it was to serve any more than decorative purpose, would value A at ($va * $pdq). And ideally would be annotated with the derivation of $pdq, showing whether it came from market or internal calculations. And this is how a subjective balance sheet would be constructed in my own personal armchair fantasy-finance league.

I said I had a question and this is what it is. Do people in the quantitative finance world construct balance sheets in anything like the way I’ve described? If they do, I cannot imagine any analytical distinction between credit risk and market risk. If they don’t, what else do they do? And why don’t they do it the way I’ve just described? Either way, as I’ve said, I am confused. And I thank anyone who can enlighten me on the subject.

My worry is that, if A is an obligation of a major bank, futures exchange, etc, and they cannot get a value (eg, via credit spreads) for $pdq, people may just be setting it to 1. And then letting their risk management or compliance people or whatever look over their positions to make sure they look okay.

If this practice is common, I can say very confidently that I consider it pernicious. It is cargo cult finance in a nutshell. It interrupts the process of rational calculation and makes balance sheets inaccurate from a subjective perspective, which is the only perspective that should matter for decision-making purposes. But of course, if it’s nonexistent or rare – no problem.

Now let’s get back to risk. When you say that:

> If any investment bank collapses because of a *any move* in the market,
> then their risk management systems were broken and they deserve no sympathy.

Let’s assume some bank, B, has a non-broken risk management system.

Are you claiming, (a), that there is no combination of events excepting acts of God or force majeure that could cause B to default?

Or are you claiming, (b), that there is no combination of events whose probability is realistically meaningful that could cause B to default?

Or are you claiming, (c), that there is no combination of events that the US and other major government, given their interest in stable financial markets and their notorious helicopter-borne strike force, can be expected to prevent that could cause B to default?

The Bank of Amsterdam, for example, could claim (a). So can the gold and silver ETFs, GLD and SLV. So can Paul Tustain’s service, and so can James Turk’s. (Disclaimer: I hold GLD and SLV, and maintain a GoldMoney holding. I have no connection with the Bank of Amsterdam, though I wish I did.)

My belief is that, for investment banks such as Goldman Sachs, you are claiming (b).

I assert that (b), in practice, amounts to (c).

(b) amounts to (c), I believe, because it amounts to analyzing individual risk events as independent probabilities, while ignoring the elephant in the living room. Which is that in the absence of the Greenspan put, pretty much all your really major serious wave-generating events would have to be assigned a probability correlation that was pretty near 1. Unless you claim (a), bank B will not be able to resist this uberperfect storm.

But in practice, since the Greenspan put does exist, the risk of the uberperfect storm is 0. Or more properly it is transferred into dollar risk. If your institutional performance is measured in dollars – a big if, about which more later – this risk can safely be ignored. Although, more on this in a sec, it is extremely unwise for anyone to actually recognize this fact.

Or in other words, you can take your own private risk and offload it to the dollar’s risk. As a result, folks on Wall Street are making enormous amounts of money by designing ever more efficient instruments to capture reward at the expense of risk. With no political changes at all, the implicit guarantee of stability that the “Greenspan put” provides tends to increase continually in value. This has fueled an unprecedented spate of credit expansion. Whether this feedback loop will remain controlled or metastasize into an actual hyperinflationary panic is beyond me, but I really do not consider it alarmist or exaggerated to diagnose it as a precancerous lesion.

This risk is not just being loaded onto the back of the dollar. Considering the normal political incentive to avoid exchange-rate fluctuations, and the unlimited power of devaluation which all CBs hold, it is loaded onto the entire fiat currency system. Holding RMB is hardly an escape. Because of official-sector interventions, there is no reason to think that any market price in the real world today presents an accurate picture of dollar risk, or anything like it.

So why do the people at Goldman Sachs not see eye-to-eye with me? Are they stupid or dishonest? I would never in a million years make such a claim. I am sure the median GS employee is, if anything, smarter than me, and that’s without excluding custodial staff.

I assert that the reason GS people do not think of (b) as (c) is, rather, that they have a tremendous rational interest in maintaining an internal corporate culture which encourages the usual banker level of impeccable probity, and prevents any gap from emerging between the appearance of honesty and its reality.

(c) amounts to a very close, and I believe (more on this in just a sec) corrupt, relationship between the financial industry and the state. It is not in either party’s interest to acknowledge or even understand the fundamental nature of this bond. The dependency is entirely mutual. It is a true symbiosis.

The reason (c), the Greenspan put, can be described as corrupt, is certainly not that there is any reason to question the probity or honesty of anyone connected with it.

The reason it’s corrupt is that it’s inconsistent with the conventionally understood goals of the organizations we know as democratic governments. Rather than contributing to prosperity, it contributes to inequality. Rather than producing stability, it produces instability. Rather than promoting productive enterprise, it promotes unproductive activity. Rather than controlling inflation, it creates inflation.

Look at all these unsubstantiated assertions in a row! But what I am stating is just the Austrian theory of the business cycle (ABCT). If you believe ABCT, and you identify the Greenspan put as a new instance of the traditional role of states in protecting banks which issue more liabilities than they can collectively redeem, as I’ve argued, you agree with these assertions. If you question ABCT, there is plenty of material on the Web to disabuse you. Again, I recommend de Soto.

This corrupt relationship affects the economics of economics itself. Since both sides of the symbiosis have an existential incentive to perceive it as healthy, and since both happen to employ a heck of a lot of economists, the conflict of interest inherent in (c) produces an artificial and equally corrupt incentive that favors schools of economics, such as the 19th-century Banking School, which fail to recognize the pernicious nature of centrally orchestrated mispricing of risk. And, of course, disfavors the economists who do, the Austrian School in the Misesian tradition.

Now, you mention gold.

I agree 100% with this observation of yours:

>The “discovery” by GATA that governments are major players in the gold
> market is funny, because that is like discovering that the Pope is Catholic.

Totally (and pretty funny itself). I would not spin it at all this way. I have a number of differences with the communication strategy GATA has adopted.

I think the main discovery of GATA is that senior officials of the US government, who can be expected to have an accurate estimation of the risks and benefits of their actions, have chosen to lie to Congress on the subject. This of course again can be taken as simply further evidence of Papal papism. But I find it a fairly unambiguous acknowledgement of the strain on the system.

The original GATA complaint, which was dismissed for technical reasons, still makes interesting reading.

You point out:

>Also, long before governments impose capital controls, they’ll start selling off
>their gold reserves to make the price of gold collapse. (Which is what the Fed did
>to silver in 1980?)

True. But this depends on two assumptions, both of which are questionable.

The first assumption is that they actually have said reserves. I think it’s fair to conclude that the CBs still have quite a substantial amount of gold. But, as Hidetoshi Tanaka of the IMF points out, this is not public information – nor is it about to be. The gold reserve figures that CBs do release are meaningless, because they generally combine gold and gold receivables on a single line. (Except, again, for Portugal in 2001. Yay, Portugal! Viva o FC Porto!)

The second assumption is that they can mobilize the reserves in a nontransparent way. For many large reserves this is impossible, or at least legally impossible. Mobilization of the US gold reserve, for instance, technically requires an act of Congress. Publicly announcing that Fort Knox is about to be poured into the 21st-century equivalent of the London Gold Pool is not exactly an effective way to suppress the price of gold.

Also, as an aside, I should note that while the central banks still do have plenty of gold, they are pretty much out of silver.

But I think the most important misconception about gold – yours, or anybody’s, really – is stated very crisply and concisely when you say:

>Actually, what I think is more likely is that the price of gold spikes,
>people really wonder if gold is worth $2000/ounce, and buy accordingly.

Or, presumably, sell accordingly.

I was addressing this issue in my last post, in the paragraph where I discussed the difference between gold and other commodities, such as oil, corn, pork bellies, etc. Let me incorporate that bit by reference. It was short and in parentheses and I can’t blame anyone for missing it, but I’d rather not retype it here, as I have already abused RGE’s hard disk enough.

Let me take a slightly different angle, and simply analyze the supply and demand relationships which control the exchange rates between gold and other commodities or currencies.

The primary source of demand for gold in the world today, I believe, is the Mengerian demand for its service as a medium of exchange. In other words, people exchange (E1) goods for gold predominantly because they intend to later exchange (E2) the gold for some other goods, rather than because they intend to use the gold itself directly.

Due to the overhead of gold trading, other goods, such as dollars, or whatever is your local currency, satisfy Mengerian demand more effectively if the period between E1 and E2 is short. But if the period is not short, and the exchange rate between gold and other goods can be expected to move in gold’s favor across it, gold is more effective.

Thus we can say the primary demand for gold in the world today is as a medium of saving.

Of course there is some industrial use of gold, but this is minor. And there is gold jewelry, which is by no means minor. But even most of the gold sold as jewelry in non-Western countries is a medium of saving. It is marked up very little and can be melted down very cheaply. After the tsunami in Indonesia, for example, I recall hearing (sorry, I don’t have a link) that there was a substantial outflow of gold from the area, as one would expect if savings were being mobilized.

The supply of gold is the 100 to 150,000 tons of the stuff that have been mined in the course of human history, and not lost in hurricanes of the Florida Keys or whatever. Another 2500 tons or so are mined every year, but this is a small change and has only minor effects – or at least, would in a theoretical free market have only minor effects. And finally, it is important to include gold that has been identified but not mined, maybe another 50,000 tons or so. Unmined gold is substitutable to some extent, in the form of gold producer equity, with actual gold. However, many people who demand gold consider this chain of title to be too long and insecure for ordinary peaceful dormition, and unmined gold is discounted substantially.

So what does the price of gold depend on? Well, it depends on the supply. Which is basically constant. And it depends on the demand. Which depends on its effectiveness as a medium of saving. Which depends on – the price of gold!

This is the famous Austrian circle. At least in its status as a good which satisfies Mengerian demand, gold has no fundamental value. Goldbugs make this error, if anything, more often than non-goldbugs.

This feedback loop has no intrinsic bias. It can be either a vicious or a virtuous circle. It was a mostly vicious circle beween 1981 and 1999. It has been a mostly virtuous circle since 1999. Of course, if you work at the Fed or Treasury, you should reverse these adjectives.

Trying to derive a number for the outcome of this feedback loop, to generate some kind of precise quantitative prediction, would be the worst kind of cargo cult finance. Of course you can try to estimate it intuitively and make a guess.

But it is just a guess. It is not a hypothesis, and its confirmation or violation cannot establish or falsify the truth of any proposition. The argument I have presented here is not a matter of inductive experiment, but of deductive logic. It is based on reason and rhetoric, not “science.” This, too, is the orthodox Misesian approach.

In other words, the inputs to the gold price are (a) the continuing devaluation of fiat, which I only expect to accelerate and which may accelerate sharply in an unstable manner; (b) official intervention, which tends to counter (a); and (c) the mere popular fashion of public opinion.

And (c) is so much larger than the other terms that if you can’t quantify it, you are wasting your time. Since no one can quantify it, no quantitative prediction is feasible.

I can say that financial fashions should shift to favor gold. But I can also say that male fashions should shift to favor shaving all the way to the ear, instead of having a little pseudo-sideburn thing that makes you look like an extra on “Midnight Cowboy.” I am generally optimistic about human nature, especially in the presence of the Internet, which disintermediates communication so effectively. But who knows, in a few years we might all have to have huge muttonchops like Ambrose Burnside. Fortunately, the modern male image has not yet felt the wrath of the extrapolating mathematician.

So when you ask:

>The obvious prediction that you are making is that the price of gold on March 15,
>2009 will be two times higher than the price of gold today. Willing to make that bet?

I will respectfully decline, and I caution everyone against making any such wager – especially with the assistance of leverage.

Do not borrow money to buy gold (or silver). Unless you want to end up like the Hunt brothers, and you don’t. As a Mengerian commodity, gold, as I hope I’ve demonstrated, is volatile in a way that makes it qualitatively different from every other asset on the planet. The only good reason to own precious metals, besides direct use, is allocated ownership of physical metal for the purpose of transferring present wealth to the future. The only good reason to sell precious metals is that you have some and you want to trade them for something else.

And with that, I think I’ve left quite enough of my yellow stain on Brad’s hard disk…

Q: I said I had a question and this is what it is. Do people in the quantitative finance world construct balance sheets in anything like the way I’ve described? If they do, I cannot imagine any analytical distinction between credit risk and market risk.

Yes there is a difference. Tomorrow the Fed increases interest rates. The bond prices change, but the spreads may not. That’s market risk. If tomorrow, company X goes under or gets its credit rating changed, the bond prices for that company change, but the interest rates in the general economy don’t. That’s credit risk.

The reason those are separate is that people who spend their time figuring analyzing a company for credit risk are different people from those that analyze interest rates.

Q: Let’s assume some bank, B, has a non-broken risk management system.

I’m claiming

d) any bank can default and anyone who thinks that banks can’t default is a fool

Given enough stress any company will default. There are know ways of reducing the chance of that happening. A bank with broken risk management is not wearing their seat belts. If you do wear your seat belts, there is a chance that someone thing cause you to crash and kill you anyway, but the odds of that happening are lower than if you wear don’t your seat belts.

Also, people in risk management at investment banks (especially GS) don’t think that what they are doing will insure that GS won’t default. You can reduce risk. You can manage risk. You can’t eliminate it. The most you can do is to make sure that you don’t default because of the known stupid ways that have caused people to default in the past. But you never know.

The only people that can’t “default” is the government and that’s only because they can print money. Inflation is a form of “slow default” and every money manager has to worry about inflation. The bond says that I’ll get $1 million in fifteen years. Great? What’s $1 million in fifteen years going to be worth?

> But in practice, since the Greenspan put does exist, the risk of the uberperfect storm is 0. Or more properly it is
> transferred into dollar risk. If your institutional performance is measured in dollars – a big if, about which more
> later – this risk can safely be ignored.

No it can’t be. If you do business in dollars, the risk of a “default” in dollars comes in inflation risk. The bond says that you will be paid $30,000 in ten years. In ten years, $30,000 may be enough to buy a postage stamp. Right now, the US government is being fiscally irresponsible enough so that people think that it will “slow default” over a period of time, which is why the money is going away from dollars into something else (like I don’t know …. RMB).

If you think that the dollar isn’t going to hold its value in the long run, the logical thing to do is to borrow a huge amount of money, buy hard assets, with the assurance that the dollars you will pay back the loans will be worth a lot less than what they are now. That’s what’s happening. If you think that a bank or government is going to default, you want to borrow every cent you can from them, since they are going to have a hard time getting that money back if they are out of business.

JW,

My first question was analytical, not organizational. Let me rephrase it yet again:

When a financially sophisticated actor has a large portfolio of assets which are obligations of other companies, does it (1) go through the exercise, not just as a secondary risk management function, but as part of its normal decision-making process, of computing quantitative expected values for those assets based on the probabilities that its counterparties will default? Does it (2) use its own estimates for those probabilities, rather than market estimates, where appropriate? And does it (3), when constructing those probabilities, consider the correlated risk of a general collapse in which the music stops and every major player turns out to have far more butts (obligations for which it is responsible) than chairs (actual goods with which it can fulfill these obligations).

If the answer to any of these questions is “no,” the result is an inflationary mispricing of risk which effectively creates virtual goods. In fact, a functioning competitive market (which Wall Street most certainly is) will force every successful player in the game to have more butts than (physical) chairs. Let me explain how this works for each “no.”

For (1), you are simply overvaluing the asset. Unless you also answer no to (3), this overvaluation will be relatively minor (credit spreads between Goldman bonds and T-bills). But it is still wrong. Mispricing cannot, by definition, be rational.

For (2), you are in danger of practicing cargo cult finance. Markets are not magic. They are very good at synthesizing information from multiple incomplete perspectives. But they cannot actually create information. A market is only as good as the guesses that go into it. If everyone is depending on the market’s estimate rather than constructing and propagating their own, the market estimate is meaningless. If almost everyone is, the market is almost meaningless.

The risk is that, in the market for things like Goldman credit spreads, relatively few people are injecting information and the information they have is not very good. Which would mean that the market estimate is not very good, either.

Consider the classic jellybean experiment: show a class a jar of jellybeans. Their average guess will be very close to the number of beans in the jar. But if you don’t actually show them the jar, just tell them you have a jar and ask them how many beans are in it, the result is quite different.

But both (1) and (2) are relatively minor effects. (3) is the real kicker.

When you write:

>d) any bank can default and anyone who thinks that banks can’t default is a fool

I wish you would look back at my previous post and see the list of “banks” present and past that can’t, except as a result of acts of God or force majeure, default.

Of course, all that means is that we mean different things by the word “bank.”

You mean a lending bank in the English fractional-reserve tradition. Austrians argue that this tradition is, like many historically venerated traditions, pathological. My specific point is that the increasing power and flexibility of computerized free-market finance is increasingly incompatible with this ancient pathology, and that our present-day financial risks and imbalances seem to fit very neatly into its historical pattern of instability and cyclical bubbles, in a way that suggests greater danger than ever.

In any case, your faction rules and it owns the word. So I’ll give it to you. A “bank,” let’s say, is any financial business which balances its balance sheet with assets and obligations whose risks and maturities are mismatched. So, by definition, you are right: any bank can default. Institutions like the Bank of Amsterdam or GoldMoney are monetary storage services, not “banks.”

Randall Kroszner and Tyler Cowen are certainly not Austrians. At least, I wouldn’t call them that. I would call them Friedmanites with perhaps a mild sprinkling of Hayek bits. But I believe Friedman himself occasionally mused favorably about 100% banking – albeit in a vaguely elegiac old-country mood, with nothing like the ruthless hatred of a Murray Rothbard – and this old paper of Cowen and Kroszner’s is a good example of how mass-market consumer investment should work in a sound-money environment. (Abstract: a mutual fund can be used as a checking account.)

The fundamental problem with all banks, and the reason that orthodox Austrians believe they should cease to exist in their present form, is that they profit by concealing information from the market. If a bank’s asset portfolio were public information, as a mutual fund’s (mostly) is, holders of its obligations would be able to accurately quantify their expected value. Since this information is not, in fact, public, the situation is that of the unseen jellybean jar.

Consider a classic 19th-century wildcat gold bank which issues more notes than it has gold. How should we value these obligations? If we value them at par, and the facts of the case are that the bank has issued ten notes for every bit of gold, and it holds no other assets, we are simply validating a fraud.

If the bank does have other assets, and they are securities, loans, etc, we cannot sensibly value the notes without valuing the assets. If they are IOUs from the CEO’s cousin Ricky, who is doing some innovative work in Florida swampland, we may still have a problem.

The fact that the bank’s portfolio is a closely held secret, and that any time someone turns up and demands gold in exchange for notes he gets it, does not alter this reality in the slightest.

If you accept obligations from an actor whose finances are nontransparent, whether you value them at par, or whether you discount them using information that is derived from a hidden-jellybean-jar market, you are making a fundamental mistake.

The only reason that rational and intelligent private actors make this mistake is that actors with fiat power, such as the US government, make it profitable for them to do so.

When an actor with fiat power intervenes to insure banks, or other financial market players, the effective result is monetary dilution, as liquid instruments are overvalued. Mises’ terminology for this effect is the expansion of circulation credit. But some people still just call it “inflation.”

So when you write:

>> But in practice, since the Greenspan put does exist, the risk of the uberperfect storm is 0. Or more properly it is
>> transferred into dollar risk. If your institutional performance is measured in dollars – a big if, about which more
>> later – this risk can safely be ignored.
>
> No it can’t be. If you do business in dollars, the risk of a “default” in dollars
> comes in inflation risk. The bond says that you will be paid $30,000 in ten
> years. In ten years, $30,000 may be enough to buy a postage stamp.

I fear you are being slightly disingenuous in your selection of context. I think the charge that I’m ignoring dollar risk is a difficult one to make if you read the entire conversation. All I meant was that when you make quantitative financial calculations in dollars, these calculations, by definition, do not consider dollar risk. It doesn’t mean you can’t use the result of these calculations in another calculation which does consider dollar risk.

Though of course “dollar risk” is a poorly defined phrase. The risk is in fact the opportunity cost of the reduced result you obtain by using dollars, rather than some other good, as a medium of intertemporal exchange (ie, a medium of saving). Dollar risk is always relative, in other words, to some other commodity – as Big Al would say, it has no fixed frame of reference.

And when you write:

> Right now, the US government is being fiscally irresponsible enough so
> that people think that it will “slow default” over a period of time, which
> is why the money is going away from dollars into something else (like
> I don’t know …. RMB)

RMB! Whoa. I totally hadn’t considered that.

Obviously, these aren’t the droids I was was looking for. I don’t know how I ever thought they were! I mean, isn’t that weird? But anyway. I guess my new handle will have to be “yuanbug.”

So how do ya circumvent those capital controls, anyway?

JW,

You write:

> > 1) as part of its normal decision-making process, of computing
> > quantitative expected values for those assets based on the probabilities
> > that its counterparties will default?
>
> For risk management, it doesn’t care about expected values. It cares about
> worst case scenarios. The probabilities are irrelevant here. Even if the
> probablility is 0.001, a trade that will cause the bank to fold is unacceptable.

First, since the phrase preceding my quote is “not just as a secondary risk management function, but”, I’m afraid I’ll have to repeat my charge of aggressive quoting. (But perhaps you just misunderstood the word “secondary.” See below.)

Second, I thought you said any bank can default? So how is the probability 0? I feel like I am misunderstanding something here. Either it is epsilon (0.001, 0.00001, etc) or it is 0. It can’t be both.

> > The fact that the bank’s portfolio is a closely held secret.
>
> It’s actually not. They have annual reports with the sort of information you
> need to have a guess at what is going on.

A guess?

This is what I find so strange about the way Wall Street does business. One minute any probability of some event is unacceptable. The next we are using words like “guess.”

Is this a technical term? Is there some scale with standardized gradations? Is a guess, perhaps, more accurate than a whirl, but slightly less precise than a stab?

And then you say:

> 1) Banks are intermediaries, and a lot of the time you don’t care too much
> about their finances since they are just facilitating the trade.

One, I am specifically talking about liabilities of banks (and futures markets). They are hardly just “facilitating the trade.”

I also find your use of the phrase “too much” disturbing. I mean, what are we talking about, 2.5 gazillion dollars, here? If we can’t be quantitatively precise, you’d think we could at least be rhetorically precise.

Let me switch to an extremely concrete example. Let’s compare the value of a Comex gold future to the value of an equivalent amount of metal in a GLD share, Comex warehouse receipt, etc.

The future is exposed to the default risk of NYMEX. The others are not. Nor are they exposed to any comparable institutional risk. They can only lose value through acts of God or force majeure. Which are also things, of course, that can happen to NYMEX.

(Note that the default risk I describe here is the contract default risk. This is not the same as the risk that NYMEX itself will fail. It is the risk that the contract will not deliver gold as specified. For a CFTC regulated futures exchange, unfortunately, these values are nowhere near equal. For example, even if NYMEX issued corporate bonds (which I doubt, but I don’t know), the spread on these bonds cannot be used to measure contract default risk.)

Even if this risk is epsilon – which, frankly, is not exactly first the Greek letter that leaps out at me on this one – the value of the instruments which do not carry it does not include this epsilon. So they, measuring by expected weight of gold, must be worth more.

Arbitrage, anyone? Bueller? Bueller?

So here is another non-rhetorical question, which I’d be curious to hear your answer to. Given these facts, why does anyone (besides commercial users who actually expect to need a delivery in three months, or whatever) buy gold futures? Is it just the built-in cheap margin financing? Is it just because that’s how they’re used to doing it? Or is it something else that I’m missing?

> 2) You can get reasonably good information on the state of a corporation
> by looking at its accounts. Of course, they could be lying, but there are
> some systems in place to reduce (note I said reduce and not eliminate) that
> possibility.

Again, the phrase “reasonably good” strikes me as odd coming from a quant. Either you can assign probabilities or you can’t. If you can, why not use them?

A monetary gold reserve, like (the gold ETF) GLD, is not fundamentally different from any other corporation. A share of GLD is a paper promise of gold just like anything else. GLD just maximizes the value, measured in gold, of its gold obligations, by adopting a policy of complete risk avoidance and transparency. The noninstitutional DGCs, like GoldMoney and BullionVault, are incredibly aggro about this. You legally own your balance at both of these institutions. And there is basically no corporate information about either that is confidential. They don’t post their internal email logs on the Web, for example, but maybe they should.

In a free market, I believe, corporations that stand for obligations, whether something very simple like a gold delivery, or something crazy like a CDO, would always profit from transparency, because people would arbitrage the epsilons of their default risk and shift to the least risky counterparty. No obligation would be valued at par, with zero epsilon. Issuers with only force-majeure risk would always beat companies with market event default risk in the market for equivalent instruments.

My suspicion, from which you have not (at least yet) dissuaded me, is that state intervention in low-probability events prevents this market from existing in the quantitative sense I have just described. Instead, risk management is a form of top-down planning, with the top being Basel 2 or whatever. Since, as an Austrian, I believe markets are more effective than planning, it should not be a surprise that I prefer risk markets to risk planning.

> What I’m telling you is that the market says that there is a
> very good chance of a “slow default” in dollars, and that is
> why people are trying to get out of dollars and into gold, Euros,
> RMB, Yen, real estate, anything else other than dollars which
> people are trusting less and less.

If you still suspect that I am unaware of this, we definitely have a failure to communicate.

However, I am not sure you have assimilated the fact that the issuers of euros, yen, and RMB have both the political motivation and the institutional instruments to minimize deviations in the exchange rate between their currencies and the dollar. Therefore, it is hard to understand why anyone would use these tools to speculate on a dollar collapse. In my opinion, when BWII goes, it is more likely to be a fiat collapse than a dollar collapse. Of course, this is just my own opinion. The question is a political one. It is not at all a matter of economics.

Both motivation and instrument exist for both gold and real estate, but in very different forms. Real estate is devalued by overconstruction. Some markets have very secure scarcity, but others cannot be expected to deviate sustainably from production cost. A 10x increase in the gold price might double or triple new production within a few years, but new production of gold has only a minor influence on supply. Gold, therefore, is simply hard to devalue.

However, governments would very much prefer that their citizens use real estate, rather than gold, as an instrument of saving. Once they start using gold as an instrument of saving, they may get the idea that they could use it as an instrument of spending. Which would eliminate the governments’ ability to finance themselves by debasing the currency. Etc, etc.

So there are many, many, political advantages to residential real estate in scarcity-priced markets as an instrument of saving. And there will always be. (Not to mention that a house, unlike gold, is actually useful outside of its exchange value.)

I’m not so sure that the unwinding hasn’t begun. I have considered the dollar support to be an artifact, in part, of a complex, fluid process of competitive devaluation. The mercantilists are devaluing faster than the FX market (and US government) are devaluing the dollar. A second part is that much of the currency inflation is muted by trade expansion at a rate sufficient to soak up the excess currency as “working capital”. The game unwinds when export growth slows, which has begun but is not yet critical.

I expect the Chinese response will be deflationary as they crank up the export machine in a frantic, no holds barred attempt to keep the game running.

The restoration of fiscal and monetary sanity in the US will bring about a change in the new world order. Hopefully, it is orderly.

N Roubini has been accurate in both time and quantity for the past year. Calling the top or bottom of an irrational bubble is by definiton, impossible.

Best regards