It is time to stop quivering in our boots in pointless fear of the future and just roll up our sleeves and build it.
- Ray Pierrehumbert

Monday, September 29, 2008

It's Raining and It's Windy and I'm Sad

The rest of the cartoon (H/T Dot Earth) is here. Another timely toon here.

H/T Atmoz for the title of this posting.

The NY Times points to an audio production of an outsider's asking stupid questions about finance. It promises to be very interesting.

I thought I'd capture my opinions about what I'll call
the wtf, by which I mean the current financial/political situation, before listening to that report.

My background, as regular readers know, is a PhD in climatology and a long standing interest in the processes of policy and collective decision making under complexity. So of course the present clstrfk fascinates me.

My opinion is informed by an undergrad class in economics in the 70s taught by a Keynesian, occasional and spotty reading in the area since, and fundamnetal doubt regarding some axioms of economic thought, especially that growth in measurable financial transactions is a useful measure of public well-being. The last keeps me very separate from Keynesians like Krugman to whom I am otherwise closest. It puts me closest to a small camp of anti-growth economists, but they also seem wrong to me on technical grounds I won't get into here. I make no claim to being an expert; I merely doubt that there is much expertise at all.

For instance, here is Kalinda Stevenson, Ph.D., financial adviser, life coach stating "However, the bank cannot loan out all of its deposits. If you deposit $1,000 in the bank, the bank loans most, but not all, of your $1,000 to other customers." As I will explain briefly below, this seems obvious enough but it is completely wrong. Anyone believing this falsehood will find it impossible to understand wtf, their eyes will glaze over, they will sigh, and they will go with their "gut" which will steer them wrong. OK, her PhD turns out to be in protestant theology, so I'll put my Ph.D. up against hers as a basis for thinking about economics any day, but she doesn't stress this at all and happily poses as an expert on money. Her implied perspectives in the latter article should fascinate those who don't understand the "conservative", "red state mentality". But I digress.

Anyway, here's my own first attempt to explain wtf before shuffling off to my day job this morning.

First of all, the intrinsic worth of a dollar is obviously zero; a dollar is simply a measure of how much purchasing power everybody agrees you have. The fact that it's disconnected from gold is pretty secondary. The practical uses of gold are real enough, but the value of gold is also pretty much dominated by convention and not by actual utility. The number of dollars in the world (I use the term loosely to include all fungible currencies here) is not the total of printed and minted currency; the existence of printed and minted currency is a bit anachronistic. Most dollars are digital. So what prevents the bank from "lending" more of these bits than they actually have?

Well, in fact, nothing. More to the point, banks are encouraged to do exactly that. Lending money you don't have is the prerogative of a bank. Banks are the institutions which are licensed to lend money they don't actually have! This is the keystone of the whole situation. Currency comes form the mint, but money comes from the bank. Anyone with a license to lend money they don't have is a bank, any bank is licensed to lend money they don't have.

The crucial aspect is the multiplier, the ratio of how much they lend to how much they have.

What controls the multiplier? Could a bank abuse this and create infinite amounts of money? Well, in the past this would cause a run on the bank. Everyone with any real money on deposit would "pull it out", ultimately turning it into currency, stuffing their mattresses, and buying up ammunition. (This is of course rather silly, because even the stuff in your mattress is ultimately imaginary.) The amount of actual cash on hand at the bank would become negative, the ratio would become infinite, and the bank would find itself unable to get currency and would "collapse", leaving remaining deposits worthless. Something like this happened in 1929 if I understand correctly.

To prevent this, we have a federal reserve bank, whose function is primarily to insure the first $100K of any deposit at any bank. (Does bank consolidation limit the amount of money you can insure?) In order to insure this, the "Fed" is empowered to regulate the banks, including setting the limits on the multiplier.

This worked out brilliantly for a long time, almost 80 years in the American context; lots of "growth" happened, much of it actually corresponding to well-being. Some of us have been worried that this would eventually blow up, since many types of "growth" aren't actually improvements. Indeed, people have of late become tired, angry, stressed and shallow. This certainly feeds into our other problems. Growth addiction prevented the substitution of long-term sustainable energy supplies for short term ever-burgeoning fossil fuel consumption, and this finally provided the trigger for the wtf.

In my view, though, the wtf occurring now is decades early with respect to resource limits. Normally we could have worked through the petroleum shortfall using the genuine creativity and competence that capitalism unleashes. This ought to have been a glitch. The Kunstler scenarios never rang true for me. If we are going to be reduced to an Argentine-class collapse anytime soon, (and it looks all too likely that we are) it will be a mistake to attribute it to resource limits. Eventually those will bite, but this is too early.

Anyway, this central role of the Fed means that the libertarian "government-out-of-the-economy" advocates are thus fundamentally either ignorant or dishonest. The entire structure of modern capitalism is entirely based not only on regulatory powers but also on deliberate manipulation on the part of governmental entities. The idea that regulation could go away and everything would continue to work out is simply at odds with reality.

And this idea, the idea that markets need no regulation, the idea that modern markets even exist in the absence of regulation, not oil, not housing, not outsourcing, is the root of the problem.

After decades of republicanism, interrupted only by a brief interlude of Clintonesque "market-friendly" democratism that is not substantially different in this regard, the boundaries between banking and non-banking enterprises blurred. Because banking is so profitable compared to, you know, actual work, anyone who could get in on this something-for-nothing business looked for ways to do so. The availability of vast computing power contributed. Many brilliant people who might have been doing useful work involving derivatives and integrals were diverted to parasitical efforts involving derivatives and options. Essentially these "products" were bets on bets about bets on bets. These products essentially replicated the banks ability to manufacture money without being subject to regulation.

Remember the science fiction story about so many volumes of cross references and indexes that the actual information got lost? "Ms fnd n a lbry" it was called... It's sort of like what has happened. There were so many bets and bets about bets that they amounted to a huge financial structure dwarfing the real economy on which they were perched. And now the whole structure is tipping over.

For years, rewards went to people making elaborate bets far more than to people doing real work. (Even our heroes at Google and Apple are fundamentally in advertising and fashion, though to be sure they have done more real work than most beneficiaries of the current period!) This fed the building boom, feeding the real estate boom, feeding into many of the bets that suddenly went sour at the first provocation.

Of course, there's also a lot of resources lost to literally blowing things up these days too. But I don't think the awful tragic waste of the Iraq escapade is crucial.

My idea of what happened is that the bets about bets about bets became so huge as to so dominate; that their consequent effect is essentially to undermine the whole concept of money. More money changed hands in the phantom economy every day (if I recall correctly) than was transmitted in real transactions in a year. The bets on bets came to dominate finances. So at the first significant setback to the real economy that all these bets were magnifying, multiplier effects started cascading up the betting chain, and now they are set to cascade back down.

Because the system was unregulated, it became too risk-tolerant and brittle. At the first provocation (peak oil is not peak energy, folks) rather than adjusting it started to crack.

The $700 bn can be looked at as an effort to grout the cracks. Will it work? I don't know. Is it a good idea? Well, since the money is sort of an imaginary quantity anyway, and if we don't risk it, it will become sort of worthless anyway, I figure, yeah, it's better than instability.

For once, I have a little bit of sympathy with the extreme republicans. They say that a core tenet of capitalism is letting failures fail. Maybe we ought to take them up on it and see where that leaves us. One thing is certain; rich people would lose more than poor people would. There's a certain rough justice in that. The outcome though would be a huge leveller, something I think the hard core Republicans (representing mostly relatively rich people in relatively poor regions) would not care for.

I also think the liberal impulse is easy to understand. This looks like the middle class bailing out the rich!

The thing is, it's all paper. If the value of the paper goes away, the middle class is even more thoroughly screwed. So they really have got us over a barrel. I am not smart enough about these things to have much to say about the details, but I am smart enough to understand the political calculus of it. Pelosi is right to insist that the Democrats not be saddled with the blame; I have no issue with that. I think we should try to keep the existing mechanisms wheezing along as best we can rather than trying to reinvent them from scratch, though. So I don't know what should be done, but it seems to me like it's not nothing.

In the end, though, it's the market libertarians' incapacity to see that the system we have set up is an elaborate artifact, not a fact of nature, that is at the root of the problem. We are now in a position that the whole of capitalism needs to be reconsidered and largely reinvented, and it's hard to see who is on the scene to do the thinking.

In the short run I think we need to try to flex rather than breaking the whole thing in one swoop, though there is a strong argument for getting the collapse over with based on the plausible idea that any bailout won't work anyway. The way I see it, if things are that bad it hardly matters what we do though. It's hard to understate the risk associated with total breakage, so that's why I'm for at least trying to patch it together.

It's interesting that it's the deregulators in congress who caused the problem and thus created the risk of collapse seem to be the ones trying hardest to finish the job...

Interesting times, either way.

Update: Unsurprisingly, the Texas delegation has been particularly unenthusiastic about the bailout, this despite the extent to which Phil Gramm, former Texas senator, was a crucial player in inflating the bubble in the first place. Did you expect a population that keeps electing Ron Paul to be very enthusiastic about this thing? What's perhaps a bit surprising, though, is that the eight members from Arizona, evenly split between the major parties, were unanimously in opposition, despite the urging of their own Senator McCain.

Update: James Galbraith:
Despite the common use of language, the capital cost of this bill does not involve "taxpayer dollars." It authorizes a financial transaction, exchanging good debt (U.S. Treasury bills and bonds) for bad debt (the "troubled assets"). Many of those troubled assets will continue to earn income for some time, perhaps a long time. The U.S. Treasury commits itself to paying the interest on the debts it issues. The net fiscal cost -- which is also the net fiscal stimulus -- of this bill is the difference between those two revenue streams. Given the very low rate of interest presently prevailing on Treasury bills, this is likely to be somewhere between $20 billion per year and zero from the beginning, even if the Treasury were to issue all $700 billion in new debt at once. It is a mistake, in short, to count the capital cost as a "cost to the taxpayer."

This is not the war in Iraq. In the longer run, of course the Treasury will incur capital losses on the assets it acquires. The entire purpose of the bill is to overpay for bad assets, so as to give financial institutions a chance to recapitalize themselves.

H/T 3Quarks.

Update: Tom Friedman is apocalyptic.

Update: In the comments, David Benson agrees with me that the newfound availability of vast computing power is a component of the financial fiasco. But HPC Wire (the newsletter of the high performance computing community) takes exactly the opposite perspective!
"Why didn't the sophisticated, computerized pricing models that Wall Street firms use to predict returns and risk for complex derivatives save them from the sub-prime mortgage mess? The short answer is: Fund and portfolio managers rarely use them." Crosman goes on to reveal some problems with the algorithms themselves, noting that "some models for analyzing mortgage-backed securities don’t include house prices, which are a fairly important piece of the puzzle." In other cases, the models were simplified for the sake of expediency. One quant noted that "[t]raders will like a light model because they don’t need heavy routines that will take forever to run on their machines."
Update: Krugman not only is apocalyptic, he uses the word "apocalyptic".



Aaron said...

FDIC regulated banks ARE limited in how much they can lend. The limits are related to total assets rather than to cash. So, yes they can loan more cash than they have on hand, but the ratio of loans to assets is strictly controlled.

You could use your good credit to borrow more than your net worth and then loan it to your freinds. An FDIC bank can not. And, FDIC banks have tended to do better than say, investment banks.

The FDIC banks that got into trouble were those that were in the housing loan business as it was deregulated ("Bush's ownership society") and they got greedy. Or, they went into the deregulated housing loan business because they were greedy. This was a matter that they were not repaid for loans that they made, rather than the fact that they loaned more money than they had. Their loans were imprudent rather than excessive.

The FDIC rgulattions are a good read for any budding economist. (, When I worked for a bank, they were only available as leather bound volumes in the VP's office.) The great read is to see how they have changed over the last 8 years. "Streamlining the regulations to save money" has cost us $billion$.

tidal said...

One short comment as I muddle my way through this wtf as well...

W.r.t. "Growth addiction... In my view, though, the wtf occurring now is decades early with respect to resource limits. Normally we could have worked through the petroleum shortfall using the genuine creativity and competence that capitalism unleashes. This ought to have been a glitch... Eventually those will bite, but this is too early."

What disturbs me is that key resource limits - even IF decades away - may NOT be far enough into the future that they won't have serious impacts on our economies and capital markets in the rather near future.

Here's why. As Peter Bernstein says: "Financial markets are a kind of time machine that allows selling investors to compress the future into the present, and buying investors to stretch the present into the future." So, when you invest in a stock or a market trading at, say, 20x earnings... What you are in effect acknowledging is that 5% of the value of the investment is represented by next year's earnings... and the other 95% is represented by "something else"... The "something else" is first and foremost "the future" and our expectations about that future...

So, capital markets valuations are finely-tuned to our expectations about the future... As a thought experiment, imagine that the investment above was expected to grow at 4.5% for the next 20 years or so... and thereby theoretically justified the market's assignation of 20x multiple of next year's earnings... But suppose the prevailing market paradigm for expected growth in general were to relatively shrink from 4.5% to 3.0%... The future earnings pool would be cut by roughly 16%... But further, the investments would no longer be worth a 20x multiple... So that too would have to adjust to the lower-growth paradigm... say to 14x... What this would imply is as follows: Say we initially had a "market" that was expected to earn $5 next year, and was trading at $100. Just the assumption of reduced growth expectations for the future could easily result in a >>40% decline in the market capitalization of our enterprises NOW. Not someday in the future... NOW.

The finance math is a bit of an oversimplification there, but the magnitude and timing of valuation changes are not off materially.

Leaving aside what slower realized earnings growth would mean to incomes, entitlement program funding, trade balances, etc. as growth slowed due to the arrival of real resource limits... what concerns me more is the impact that shifts in growth assumptions can have on market caps NOW. That store of wealth is in turn critical to our abilities to fund investment in clean and sustainable energy infrastructures, etc.

And I don't think this is at all tangential to what is going on in the financial markets presently. Even if I agree that there is no direct connection in the financial contraction and resource limits, what we are witnessing is a manifestation of the same valuation contraction that I described above. The "2+2=5!" paradigm of 40:1-or-25:1-or- whatever-leveraged f.i.'s has proven to be a mirage, and part of what is happening is a compression of earnings expectations and earnings multiples. Same dynamic. In this case, the spillover is enough that we can have wealth contractions across the board, beyond just financials, or $1.2 trillion just yesterday! Consider what happens when we collectively begin to muse "Ya know, what with oil peaking, and the oceans dead with the acidity of a can of Coca-Cola, and the Arctic ice free, mebbe we should ease off on the accelerator a tad..."... And I know I am not alone in suspecting that the financial crisis is the beginning of a longer introspection of the role of government in (un)regulated markets, and the sustainability of a variety of growth models - not just in finance...

Another disquieting parallel from the financial crisis is how the thing escalated to apparently near-catastrophic proportions before we acknowledged the gravity of the situation and the required mitigation/adaptation...

I am not doom and glooming it... I think the grow-grow-grow paradigm is so heavily invested in itself that we will not see an abrupt shift. But some kind of downshift in material throughput is coming, whether gradual or abrupt (because some resource/sink limits are simply non-negotiable)...

In my most hopeful scenarios, we actually end up with the kind of near-term economic boom that some forecast can result from a massive retooling and buildout of long-life energy and resource management infrastructures... and a refocussing of economic goals to the more nebulous "development" rather than "growth" that the ecological economists call for... but I get an unnerving sense that these problems kind of sneak up on us, and we can get major discontinuties... and that "resource limits" will hit financial markets decades before they "really" hit... wtf indeed...

Dano said...

There have been two interviews on Moyers [1., A. ]that I think really get at what is going on here. 1. is Kevin Phillips and A. is Andrew Bacevich. Please take the time and watch.

That they echo what I (and Dano) have been saying for years is - IMHO - a testament to stepping back and looking. Not owning a TV helps, but these are excellent contextual backgrounds for what we have today.

Lastly, IMHO, the world isn't going to end tomorrow. That we don't have growth! growth! groooooowth!!!! for a little while - against the alwaysgrowth! mantra - won't kill us.

The ululating, gnashing of teeth and rending of garments is almost all coming from the FIRE sector and esp Finance and Insurance. Their primacy (echoing Phillips and even Kunstler) got us here in the first place.

It won't be easy to get us off our addiction to paper wealth, but we can do it.



David B. Benson said...

The availability of vast computing power contributed.


Richard said...

Excellent post on a fascinating (and unfortunately vivid) subject.

I heard the first half of that "This American Life" episode while driving back in May, and kept listening to the end, even after I parked.

Transcript here:

Epic sequence:

"The global pool of money. That's where our story begins. Most people don’t think about it but there’s this huge pool of money out there, which is basically all the money the world is saving now. Insurance companies saving for a catastrophe, pension funds saving money for retirement, the central bank of England saving for whatever central banks save for. All the world’s savings.

Ceyla Pazarbasioglu: It's a lot of money. It's about 70 trillion."

Which had to find a place to go, just when the rising US housing market, cheap credit, and shoddy 'mortgage-backed securities' seemed inviting. That, plus the bankers leveraged themselves up to 30:1 or more, to extract even more cash from bigger deals.

Also: making bets with borrowed money is great, as long as you win the bets.

Two personal recollections:

re: your observation on --
"brilliant people who might have been doing useful work involving derivatives and integrals"

My closest friend from the science high school I went to in NYC has a PhD in astrophysics (Princeton, U Chicago). He went into finance in the early 90's after eating ramen as a post-doc at Yale for a few years. He now runs the 'risk' desk at Morgan Stanley, where he's a managing director. Making -- or assessing -- models about derivatives is part of his job, I think. His work as a scientist involved making models about the first three seconds of the universe.

The pay scale in NYC jobs in finance was, until this year, 4:1 over the average pay in NYC.

The average pay at Goldman (clerks, IT people, security guards, traders) was $ 622,000 in 2006.

Because of the pay, over the past decade and more about 40% of the US's best college students (seniors at the Ivies and MIT, for example) entered the financial industry. Perhaps there has been a double cost: the perpetuation of a Ponzi scheme, and a brain drain. The talent drain may have been particularly acute because of the consuming nature of financial jobs -- with no actuators but greed and fear, & no end product but the firm's profit. 15+ hr days, 7 day weeks, roped to Blackberries and screens, with little time to reflect on the wisdom of their machinations.

A story about imaginary money: in 1999 I went to Cuba, and heard stories of their bitter economic crisis in the early 90's from a stringer for the Economist. He had arrived in 1992, when the paper currency was blowing down the street with no one bothering to pick it up. (The only real currency was the dollar, which was strong.)

David B. Benson said...

Re: the update.

My informed opinion is not changed by those remarks. Too many people trust the output of computing just because it went through some algorithm.

Cutting over to ever higher performing computers and every more complex algorithms is a form of reductionism. Reductionism does not work in science and engineering; there is no reason to suspect that it does in finance.

David B. Benson said...

As evidence:

How Wall Street Lied to Its Computers

Michael Tobis said...

"Cutting over to ever higher performing computers and every more complex algorithms is a form of reductionism. Reductionism does not work in science and engineering;"

Can I say 'amen'? No?

Well, what he said, anyway.

Actually, I'm not sure I'd call this mistake 'reductionsism', but I'd very much call it a mistake, and a seductive and obviously dangerous one as well.

I am very very much and vigorously in agreement that this is not a fruitful approach. However, the bulk of the climate science community is not in agreement with us.

Phil said...

Maybe financial and economic models are like climate models?

The more accurate they get, the less people believe them.

Phil, tongue half in cheek

David B. Benson said...

See, in particular, methodological reductionism in


EliRabett said...

Stevenson is a good example of why we need regulation of ANYONE who offers financial advice. Does she have ANY qualifications? That is going to be as important as regulating banks and brokers. The current mess has its roots in a mortgage finance industry that was unregulated, with unregulated mortgage brokers, unregulated investment banks and a whole lot of fools who play financial advisers on TV