The Long Run

the-long-run-home

Niall Ferguson’s misunderstanding of Keynes led me to the question of how humans should balance the present against the long run?

It’s hard for us primates to have a real clue about the long run — the chain of events that may occur, the kind of world that will form. In the long run — the billions of years for which Earth has existed — modern human civilisation is a flash, a momentary pulsation of order imposed by primates on the face of the Earth — modern cities, roads, ports, oil wells, telecommunications and so forth built up over a little more than a century, a little more than two or three frail human lifespans.

Human projections of the direction of the future are notoriously unreliable. Professional futurists who devote careers to mapping the trajectory of human and earthly progress are often far wide of the mark. And in the realm of markets and economics, human projectional abilities are notoriously awful — only 0.4% of money managers beat the market over ten years.

As humans, our only window to the future is our imaginations. We cannot know the future, but we can imagine it as Ludwig Lachmann once noted. And in a world where everyone is working from unique internal models and expectations — for a very general example, Keynesians expecting zero rates and deflation, Austrians expecting rising rates and inflation — divergent human imaginations and expectations is an ingredient for chaos that renders assumptions of equilibrium hopelessly idealistic.

A tiny minority of fundamental investors can beat the market — Keynes himself trounced the market between 1926 and 1946, for example by following principles of value investing (like Benjamin Graham later advocated). But like in poker, while virtually everyone at the table believes they can beat the game in the long run — through, perhaps, virtues of good judgement, or good luck, or some combination of the two — the historical record shows that the vast majority of predictors are chumps. And for what it’s worth, markets are a harder game to win than games like poker. In poker, precise probabilities can be assigned to outcomes — there are no unknown unknowns in a deck of playing cards. In the market — and other fields of complex, messy human action — we cannot assign precise probabilities to anything. We are left with pure Bayesianism, with probabilities merely reflecting subjective human judgments about the future. And in valuing assets, as Keynes noted we are not even searching for the prettiest face, but for a prediction of what the market will deem to be the prettiest face.

This means that long run fears whether held by an individual or a minority or a majority are but ethereal whispers on the wind, far-fetched possibilities. It means that present crises like mass unemployment have a crushing weight of importance that potential imagined future crises do not have, and can never have until they are upon us. As the fighters of potential future demons — or in the European case, self-imposed present demons — suffer from high unemployment and weak growth in the present (which in turn create other problems — deterioration of skills, mass social and political disillusionment, etc) this becomes more and more dazzlingly apparent.

But in the long run, the historical record shows that crises certainly happen, even if they are not the ones that we might initially imagine (although they are very often something that someone imagined, however obscure). Human history is pockmarked by material crises — unemployment, displacement, failed crops, drought, marauders and vagabonds, volcanism, feudalism, slavery, invasion, a thousand terrors that might snuff out life, snuff out our unbroken genetic line back into the depths antiquity, prehistory and the saga of human and prehuman evolution. While we cannot predict the future, we can prepare and robustify during the boom so that we might have sufficient resources to deal with a crisis in the slump. Traditionally, this meant storing crops in granaries during good harvests to offset the potential damage by future famines and saving money in times of economic plenty to disburse when the economy turned downward.  In the modern context of globalisation and long, snaking supply chains it might also mean bolstering energy independence by developing wind and solar and nuclear energy resources as a decentralised replacement to fossil fuels. It might mean the decentralisation of production through widespread molecular manufacturing and disassembly technologies. In the most literal and brutal sense — that of human extinction — it might mean colonising space to spread and diversify the human genome throughout the cosmos.

Ultimately, we prepare for an uncertain future by acting in the present. The long run begins now, and now is all we have.

About these ads

Sparkassen — A De Facto Glass-Steagall?

Ed Miliband has a very good idea to break the British lending freeze:

Ed Miliband is to make his firmest commitment to a regional-based economic policy when he proposes a network of banks around the country responsible for providing capital to businesses in their locality.

The proposals, due to be unveiled in a speech to the British chambers of commerce, mark a further attempt to map out a different industrial policy, some of which has echoes of plans for a revival of city regions set out by the coalition adviser Lord Heseltine.

Miliband will say it is time to stop tinkering with the banks and recognise a wholly new system is needed.

He will say: “We do not just need a single investment serving the country. We need a regional banking system serving each and every region of the country. Regional banks with a mission to serve that region and that region alone, not banks that are likely to say no but banks that know your region and your business; not banks that you mistrust, but banks you can come to trust.”

I would not support politicians interfering with the financial sector if the British financial sector was a successful model. But the country is still hurting from its utter failure in 2008. Back then, Ed Miliband’s predecessor Gordon chose to bail out the banking system. Had the financial sector been allowed to fail, then a new model would have been forced to emerge. But that wasn’t the case. Now, politicians must take responsibility for putting the banking system on a life support system. The current government’s attempts at reform have not succeeded in revitalising the economy.

Miliband’s idea approximates the German model of Sparkassen — publicly owned regional banks:

Supporters of the local banks claim that in 2011 total loans by the Sparkassen stood at €322bn (£280bn), whereas the total loan stock of Germany’s large commercial banks was only €177bn (£153.5bn). Like Britain’s large banks, Germany’s large commercial banks cut credit during the financial crisis; lending fell by 10% between 2006 and the middle of 2011. In contrast, the Sparkassen increased lending by 17%.

On the surface, regionalisation may be helpful in that British banks have become over-centralised and disconnected from the interests of their local customers. This may be one factor that can explain why local, small and new businesses are struggling to get credit.

But this is an even better idea than Miliband may realise. Why? Because so long as the regional banks behave solely as depository and business investment institutions, and not as investment banks, insurance brokers, hedge funds, shadow banks, or proprietary traders, or any of the other highly interconnective and risky activities favoured by today’s supermarket banks — then such a system acts as de facto Glass-Steagall-style separation between the riskier privately-owned national and international-level commercial banks, and the regional level business investment and savings banks.

Such a system also echoes the recommendation made by Nassim Taleb, to nationalise the parts of the banking system that act as a public utility, and deregulate the rest so it is free to gamble, speculate, succeed and fail without significantly destabilising business lending, public savings, and the wider economy.

Taleb on Overstabilisation

It’s nice to know that Taleb is preaching more or less the same gospel that I am.

Via the NYT:

Stabilization, of course, has long been the economic playbook of the United States government; it has kept interest rates low, shored up banks, purchased bad debts and printed money. But the effect is akin to treating metastatic cancer with painkillers. It has not only let deeper problems fester, but also aggravated inequality. Bankers have continued to get rich using taxpayer dollars as both fuel and backstop. And printing money tends to disproportionately benefit a certain class. The rise in asset prices made the superrich even richer, while the median family income has dropped.

Overstabilization also corrects problems that ought not to be corrected and renders the economy more fragile; and in a fragile economy, even small errors can lead to crises and plunge the entire system into chaos. That’s what happened in 2008. More than four years after that financial crisis began, nothing has been done to address its root causes.

Our goal instead should be an antifragile system — one in which mistakes don’t ricochet throughout the economy, but can instead be used to fuel growth. The key elements to such a system are decentralization of decision making and ensuring that all economic and political actors have some “skin in the game.”

Two of the biggest policy mistakes of the past decade resulted from centralized decision making. First, the Iraq war, in addition to its tragic outcomes, cost between 40 and 100 times the original estimates. The second was the 2008 crisis, which I believe resulted from an all-too-powerful Federal Reserve providing cheap money to stifle economic volatility; this, in turn, led to the accumulation of hidden risks in the economic system, which cascaded into a major blowup.

Just as we didn’t forecast these two mistakes and their impact, we’ll miss the next ones unless we confront our error-prone system. Fortunately, the solution can be bipartisan, pleasing both those who decry a large federal government and those who distrust the market.

First, in a decentralized system, errors are by nature smaller. Switzerland is one of the world’s wealthiest and most stable countries. It is also highly decentralized — with 26 cantons that are self-governing and make most of their own budgetary decisions. The absence of a central monopoly on taxation makes them compete for tax and bureaucratic efficiency. And if the Jura canton goes bankrupt, it will not destabilize the entire Swiss economy.

In decentralized systems, problems can be solved early and when they are small; stakeholders are also generally more willing to pay to solve local challenges (like fixing a bridge), which often affect them in a direct way. And when there are terrible failures in economic management — a bankrupt county, a state ill-prepared for its pension obligations — these do not necessarily bring the national economy to its knees. In fact, states and municipalities will learn from the mistakes of others, ultimately making the economy stronger.

It’s a myth that centralization and size bring “efficiency.” Centralized states are deficit-prone precisely because they tend to be gamed by lobbyists and large corporations, which increase their size in order to get the protection of bailouts. No large company should ever be bailed out; it creates a moral hazard.

Consider the difference between Silicon Valley entrepreneurs, who are taught to “fail early and often,” and large corporations that leech off governments and demand bailouts when they’re in trouble on the pretext that they are too big to fail. Entrepreneurs don’t ask for bailouts, and their failures do not destabilize the economy as a whole.

Second, there must be skin in the game across the board, so that nobody can inflict harm on others without first harming himself. Bankers got rich — and are still rich — from transferring risk to taxpayers (and we still haven’t seen clawbacks of executive pay at companies that were bailed out). Likewise, Washington bureaucrats haven’t been exposed to punishment for their errors, whereas officials at the municipal level often have to face the wrath of voters (and neighbors) who are affected by their mistakes.

If we want our economy not to be merely resilient, but to flourish, we must strive for antifragility. It is the difference between something that breaks severely after a policy error, and something that thrives from such mistakes. Since we cannot stop making mistakes and prediction errors, let us make sure their impact is limited and localized, and can in the long term help ensure our prosperity and growth.

Who Should Be Giving Thanks This Thanksgiving?

Not the wider public.

Our financial system is broken. Our political system is broken. Oligarchs and their cronies reap easy rewards — bailouts, crony capitalism, corporate handouts, liquidity injections, favourable “regulation” (that puts oligarchs’ competition out of a business) — while taxpayers pay the bill.

But no such thing lasts forever.

Thanksgiving is very much the day of the black swan. Nassim Taleb used the example of a turkey fattened up for Thanksgiving as an example of a black swan phenomenon. The turkey sees itself being fed every day by the turkey farmer and assumes based on past behaviour that this will continued indefinitely until the day comes when the farmer kills the turkey. Nothing in the turkey’s limited experiential dataset suggested such an event.

But Thanksgiving also commemorates the end of pre-Columbian America, a huge earth-shattering black swan for the people of the Americas. The day before the first European immigrants landed in North America, very little in the Native Americans’ dataset suggested what was to come.

In a globalised and hyper-connected world, drastic systemic change can occur faster than ever before.

All it takes is the first spark.

George Osborne & Big Banks

The Telegraph reports that George Osborne thinks big banks are good for society:

The Chancellor warned that “aggressively” breaking up banks would do little to benefit the UK and insisted the Government’s plans to put in place a so-called “ring fence” to force banks to isolate their riskier, investment banking businesses from their retail arm was the right way to make the financial system safer.

“If we aggressively broke up all of our big banks, I am not sure that, as a society, we would benefit from it,” he said. “We don’t have a huge number of banks, sadly, large banks. I would like to see more.

His comments came as he gave evidence to the parliamentary commission on banking standards where he was accused of attempting to pressure members into supporting his ring-fencing reforms.

“That work has been accepted, as far as I’m aware, by all the major political parties. We are now on the verge of getting on with it,” he said.

Several members of the Commission have argued in favour of breaking up large banks, including former Chancellor, Lord Lawson.

This is really disappointing.

Why would Osborne want to see more of something which requires government bailouts to subsist?

Because that is the reality of a large, interconnective banking system filled with large, powerful interconnected banks.

The 2008 crisis illustrates the problem with a large interconnective banking system. Big banks develop large, diversified and interconnected balance sheets as a sort of shock absorber. Under ordinary circumstances, if a negative shock (say, the failure of a hedge fund) happens, and the losses incurred are shared throughout the system by multiple creditors, then those smaller losses can be more easily absorbed than if the losses were absorbed by a single creditor, who then may be forced to default to other creditors. However, in the case of a very large shock (say, the failure of a megabank like Lehman Brothers or — heaven forbid! — Goldman Sachs) an interconnective network can simply amplify the shock and set the entire system on fire.

As Andrew Haldane wrote in 2009:

Interconnected networks exhibit a knife-edge, or tipping point, property. Within a certain range, connections serve as a shock-absorber. The system acts as a mutual insurance device with disturbances dispersed and dissipated. But beyond a certain range, the system can tip the wrong side of the knife-edge. Interconnections serve as shock-ampli ers, not dampeners, as losses cascade. The system acts not as a mutual insurance device but as a mutual incendiary device.

Daron Acemoglu (et al) formalised this earlier this year:

The presence of dense connections imply that large negative shocks propagate to the entire fi nancial system. In contrast, with weak connections, shocks remain con fined to where they originate.

What this means (and what Osborne seems to miss) is that large banks are a systemic risk to a dense and interconnective financial system.

Under a free market system (i.e. no bailouts) the brutal liquidation resulting from the crash of a too-big-to-fail megabank would serve as a warning sign. Large interconnective banks would be tarnished as a risky counterparty. The banking system would either have to self-regulate — prevent banks from getting too interconnected, and provide its own (non-taxpayer funded) liquidity insurance in the case of systemic risk — or accept the reality of large-scale liquidationary crashes.

In the system we have (and the system Japan has lived with for the last twenty years) bailouts prevent liquidation, there are no real disincentives (after all capitalism without failure is like religion without sin — it doesn’t work), and the bailed-out too-big-to-fail banks become liquidity sucking zombies hooked on bailouts and injections.

Wonderful, right George?

The Problem With Centralisation

Nassim Taleb slams the European project. Perfect timing to counteract the Nobel Peace Prize nonsense.

Via Foreign Policy:

The European Union is a horrible, stupid project. The idea that unification would create an economy that could compete with China and be more like the United States is pure garbage. What ruined China, throughout history, is the top-down state. What made Europe great was the diversity: political and economic. Having the same currency, the euro, was a terrible idea. It encouraged everyone to borrow to the hilt.

The most stable country in the history of mankind, and probably the most boring, by the way, is Switzerland. It’s not even a city-state environment; it’s a municipal state. Most decisions are made at the local level, which allows for distributed errors that don’t adversely affect the wider system. Meanwhile, people want a united Europe, more alignment, and look at the problems. The solution is right in the middle of Europe — Switzerland. It’s not united! It doesn’t have a Brussels! It doesn’t need one.

The future is unpredictable. In economics some decisions will be lead to desired results and others will not. Real-world outcomes are ultimately impossible to predict, because the real world is chaotic and no simulation can ever model the real world in precise detail; the map is not the territory.

Centralisation concentrates decision-making. Centralisation acts as a transmission mechanism to transmit and amplify the effects of centralised decisions throughout a system. This means that when bad decisions are made — as inevitably happens in human behaviour — the entire system will be damaged. Under a decentralised system, there is no such problem. Under a decentralised heterogeneous system, mistakes are not so easily transmitted or amplified. Centralisation is fragile.

And central planning is mistake-prone. Central planners are uniquely ineffective as resource allocators. Free markets transmit information; the true underlying state of supply and demand. Without an open market to transmit price information, central planners cannot allocate resources according to the true state of supply and demand. Capital, time, and labour are allocated based on the central planner’s preferences, rather than the preferences of the wider society.

These two factors taken together mean that centralised systems tend to be both fragile and mistake-prone. That is a dangerous — and unsustainable — combination.

Double or Nothing: How Wall Street is Destroying Itself

There’s nothing controversial about the claim— reported on by Slate, Bloomberg and Harvard Magazine — that in the last 20 years Wall Street has moved away from an investment-led model, to a gambling-led model.

This was exemplified by the failure of LTCM which blew up unsuccessfully making huge interest rate bets for tiny profits, or “picking up nickels in front of a streamroller”, and by Jon Corzine’s MF Global doing practically the same thing with European debt (while at the same time stealing from clients).

As Nassim Taleb described in The Black Swan this strategy — betting large amounts for small frequent profits — is extremely fragile because eventually (and probably sooner in the real world than in a model) losses will happen (and, of course, if you are betting big, losses will be big). If you are running your business on the basis of leverage, this is especially dangerous, because facing a margin call or a downgrade you may be left in a fire sale to raise collateral.

This fragile business model is in fact descended from the Martingale roulette betting system. Martingale is the perfect example of the failure of theory, because in theory, Martingale is a system of guaranteed profit, which I think is probably what makes these kinds of practices so attractive to the arbitrageurs of Wall Street (and of course Wall Street often selects for this by recruiting and promoting the most wild-eyed and risk-hungry). Martingale works by betting, and then doubling your bet until you win. This — in theory, and given enough capital — delivers a profit of your initial stake every time. Historically, the problem has been that bettors run out of capital eventually, simply because they don’t have an infinite stock (of course, thanks to Ben Bernanke, that is no longer a problem). The key feature of this system— and the attribute which many institutions have copied — is that it delivers frequent small-to-moderate profits, and occasional huge losses (when the bettor runs out of money).

The key difference between modern business models, and the traditional roulette betting system is that today the focus is on betting multiple times on a single outcome. By this method (and given enough capital) it is in theory possible to win whichever way an event goes. If things are going your way, it is possible to insure your position by betting against your initial bet, and so produce a position that profits no matter what the eventual outcome. If things are not going your way, it is possible to throw larger and larger chunks of capital into a position or counter-position again and again and again —mirroring the Martingale strategy — to try to compensate for earlier bets that have gone awry (this, of course, is so often the downfall of rogue traders like Nick Leeson and Kweku Adoboli).

This brings up a key issue: there is a second problem with the Martingale strategy in the real world beyond the obvious problem of running out of capital. You can have all the capital in the world (and thanks to the Fed, the TBTF banks now have a printing-press backstop) but if you do not have a counter-party to take your bets  (and as your bets and counter-bets get bigger and bigger it by definition becomes harder and harder to find suitable counter-parties) then you are Corzined, and you will be left sitting on top of a very large load of pain (sound familiar, Bruno Iksil?)

The obvious real world example takes us back to the casino table — if you are trying to execute a Martingale strategy starting at $100, and have lost 10 times in a row, your 11th bet would have to be for $204,800 to win back your initial stake of $100. That might well exceed the casino table limits — in other words you have lost your counter-party, and are left facing a loss far huger than any expected gains.

Similarly (as Jamie Dimon and Bruno Iksil have now learned to their discredit) if you have built up a whale-sized market-dominating gross position of bets and counter-bets on the CDX IG9 index (or any such market) which turns heavily negative, it is exceedingly difficult to find a counter-party to continue increasing your bets against, and your Martingale game will probably be over, and you will be forced to face up to the (now exceedingly huge) loss. (And this recklessness is what Dimon refers to as “hedging portfolio risk“?)

The really sickening thing is that I know that these kinds of activities are going on far more than is widely recognised; every time a Wall Street bank announces a perfect trading quarter it sets off an alarm bell ringing in my head, because it means that the arbitrageurs are chasing losses and picking up nickels in front of streamrollers again, and emboldened by confidence will eventually will get crushed under the wheel, and our hyper-connected hyper-leveraged system will be thrown into shock once again by downgrades, margin calls and fire sales.

The obvious conclusion is that if the loss-chasing Martingale traders cannot resist blowing up even with the zero-interest rate policy and an unfettered fiat liquidity backstop, then perhaps this system is fundamentally weak. Alas, no. I think that the conclusion that the clueless schmucks at the Fed have reached is that poor Wall Street needs not only a lender-of-last-resort, but a counter-party-of-last-resort. If you broke your trading book doubling or quadrupling down on horseshit and are sitting on top of a colossal mark-to-market loss, why not have the Fed step in and take it off your hands at a price floor in exchange for newly “printed” digital currency? That’s what the 2008 bailouts did.

Only one problem: eventually, this approach will destroy the currency. Would you want your wealth stored in dollars that Bernanke can just duplicate and pony up to the latest TBTF Martingale catastrophe artist? I thought not: that’s one reason why Eurasian creditor nations are all quickly and purposefully going about ditching the dollar for bilateral trade.

The bottom line for Wall Street is that either the bailouts will stop and anyone practising this crazy behaviour will end up bust — ending the moral hazard of adrenaline junkie coke-and-hookers traders and 21-year-old PhD-wielding quants playing the Martingale game risk free thanks to the Fed — or the Fed will destroy the currency. I don’t know how long that will take, but the fact that the dollar is effectively no longer the global reserve currency says everything I need to know about where we are going.

The bigger point here is whatever happened to banking as banking, instead of banking as a game of roulette? You know, where investment banks make the majority of their profits and spend the majority of their efforts lending to people who need the money to create products and make ideas reality?