“We’re gonna win the game. I guarantee it.”
– Joe Namath, 3 days before Super Bowl III
“Our torments also may in length of time Become our Elements.”
– John Milton, “Paradise Lost”
“When did the future switch from being a promise to being a threat?”
– Chuck Palahniuk, “Invisible Monsters”
“Some say the world will end in fire, Some say in ice. From what I’ve tasted of desire, I hold with those who favor fire.”
– Robert Frost, “Fire and Ice”
S = k * log W
– Ludwig Boltzman’s entropy formula, as carved on his gravestone
So how does the Story end?
We are in the throes of the greatest monetary policy experiment the world has ever seen, and there are two popular Narrative arcs competing to project the ultimate resolution of the Story of QE. On the one hand we have the Happy Ending, where the Fed unwinds its $4 trillion balance sheet over time and we return to the halcyon days of either the 1990’s or the mid-2000’s, depending on your political persuasion. In this story arc the Fed voluntarily abdicates its throne as master of all things market-related in favor of its former role as a beneficial eminence grise, and we all live happily ever after. On the other hand we have the Shocking Ending, where the Fed loses control over inflation expectations and the long-end of the yield curve, and we return to the sad days of either Jimmy Carter or Weimar Germany, depending on your political persuasion. In this story arc the Fed flails around like a Mad God, leaving ruin in its wake and gold at $10,000 per ounce.
Status quo opinion leaders are enormously invested in the Happy Ending, for obvious reasons, and insurgent opinion leaders are equally invested in the Shocking Ending, for equally obvious reasons. For all I know, either one of these story arcs may turn out to be right. I don’t have a crystal ball, sad to say, and I can imagine a fact pattern that would drive the political and economic outcome to either extreme scenario. In fact, when a persuasive opinion leader talks about the path to either extreme scenario, it all sounds quite plausible to me until, of course, I hear a still more persuasive opinion leader talk about the opposite scenario.
We are, as social animals, evolved over millions of years and culturally trained over tens of thousands of years to respond to these story arcs, the more dramatic the better. It’s no accident that you find myths and cultural story arcs in every human society in every age, and you are kidding yourself if you think that mythology is any less prevalent or powerful in our modern “scientific” world. Robert Frost speaks for most of us, I believe, in his experience of desire and his preference for a world that ends in fire rather than ice, but what’s most striking is that Frost casts these as the only two options for how the story ends.
I want to suggest a third ending to the story, one that is terribly unsatisfying from a human behavioral perspective, but one that I believe is far more likely from a historical perspective – the Entropic Ending, the long slog of a gray winding-down, neither fire nor ice, neither Happy nor Shocking, where the transformation of emergency monetary policy into permanent government program creates a low growth, low inflation political equilibrium that can last for decades. Stocks will go up and stocks will go down, but not by much either way. Perpetually disappointing growth translates into persistently dashed expectations of corporate earnings growth, but the programmatic Fed backstop of financial asset prices essentially outlaws a significant price decline. There are neither secular bull markets nor secular bear markets in an Entropic Ending, just an ossification of an increasingly mediocre status quo.
Certainly there will be moments of political theatre, as we are experiencing today with the media-driven Sturm und Drang over a government “shut-down”, where we are forced to bear witness to the heartache of a couple whose wedding at Yellowstone National Park was cancelled or the tragedy of an EPA scientist who must start his research experiment all over again. Oh, the humanity! And certainly there will be moments of market exuberance, driven by, say, the thrilling prospects of yet another LTRO program or by Chinese growth coming in at 7.5% rather than 7.0%. Neither the business cycle nor animal spirits are entirely eliminated in an Entropic Ending, but they are severely muted, and both market spikes down and market spikes up should be faded in this scenario.
At the heart of the Entropic Ending is the behavioral consequence of a sea change in the perceived meaning of the Fed and its policies, from a probabilistic promise of financial asset price support to an explicit programmatic guarantee. The Bernanke market put is nothing like the Greenspan market put, not because it is quantitatively larger but because it has been signaled and reinforced to the point of Common Knowledge certainty. From a game theoretic perspective, there is an enormous difference in what Tom Schelling called “the threat that leaves something to chance” and the threat (or promise) that is perceived to be certain in its delivery. As counter-intuitive as it may seem, behavioral equilibria driven by probabilistic promises are actually more stable than those driven by guarantees, and this is why government policy guarantees (from Social Security retirement insurance to Medicare medical insurance to, now, QE growth insurance) must be backed by the full faith and credit of the US government. Without the deepest of deep pockets to stabilize what is otherwise a weak equilibrium, these programs would all end in tears sooner rather than later, creating their own versions of the Shocking Ending. With the unconditional support of the US government, on the other hand, the Shocking Ending can be pushed off for decades. But it’s silly to believe that there is not a crushing cost associated with the ever-increasing measures required to keep a policy guarantee intact, a cost that expresses itself in lower growth and fading chances of the Happy Ending. In the end, Entropy always wins, and policy guarantees just accelerate that process.
Tom Schelling, who won the 2005 Nobel Prize in Economics for his work in game theory, was less interested in macroeconomic policy than he was in defense policy. In particular, his specialty was understanding the logical underpinnings of nuclear deterrence, and in books like The Strategy of Conflicthe laid out the foundations for a US nuclear policy that both avoided Armageddon and won the Cold War. Schelling is a largely unsung hero of that conflict, a brilliant thinker, and if you could read only one book on game theory you would make a fine choice with The Strategy of Conflict.
The broad thrust of Schelling’s work on nuclear deterrence was to question the dogma of his day, Mutual Assured Destruction (MAD). MAD is based on an unconditional guarantee…if you Russians cross certain clearly demarcated lines we Americans will blow you up, even though we understand that we will also be blown up in the process. Since it’s hard to enjoy the fruits of, say, a tank invasion through the Fulda Gap when Moscow, Stalingrad, and every other Russian city is a smoldering radioactive ruin, MAD does indeed create a strong behavioral incentive not to invade West Germany. And vice versa, the US was strongly incentivized by MAD not to invade Cuba or support the occasional anti-Soviet movement in Eastern Europe, creating a behavioral equilibrium of, if not Peace, then at least No Direct War Between The Big Boys. Unfortunately, as Schelling (and to be fair, many others as well) pointed out, there are two fundamental weaknesses with MAD.
First, while it may be somewhat credible that Americans would prefer mutual suicide to a Russian conquest of the Western United States (Red Dawn notwithstanding), can you really say the same about a Russian attack on Japan? On West Berlin? The farther out you draw the “Do Not Cross” lines, the less believable MAD becomes, and that’s a bad thing because the stability of MAD’s behavioral equilibrium is entirely determined by the certainty of the promise to respond with overwhelming force. MAD depends on a guarantee that must be maintained at all costs, which is what makes it such an expensive and fragile strategy.
Second, what happens if your adversary gets right up to the clear line you’ve drawn and just pokes at it a little? He doesn’t cross it with tanks, but engages in more subtle but still entirely unwelcome advances. If you tolerate that first little poke, isn’t it pretty obvious that you will tolerate a second poke, and a third, and so on until that original line in the sand is only useful for deterring the most obvious and overt threats? Preventing a Warsaw Pact armored column from encircling Frankfurt is certainly a worthy goal, but if that’s the only behavioral equilibrium that MAD can achieve, at the cost of hundreds of billions of dollars and the non-trivial chance of human extinction if either party makes a mistake … well, that seems like a rather poor policy choice. Stanley Kubrick crystallized the susceptibility of programmatic MAD to miscommunication and poor signaling in his classic movie Dr. Strangelove: Or How I Learned to Stop Worrying and Love the Bomb, and if there’s a better visual depiction of the absurdity of a guaranteed massive nuclear response than Slim Pickens riding a warhead down to Earth I have yet to find it.
Not only did Schelling collaborate extensively with Kubrick in making Dr. Strangelove, but also (and more importantly) his work was instrumental in transforming US strategic doctrine away from MAD and towards a probabilistic, war-fighting policy. Probabilistic in the sense that a few clear lines were replaced by many fuzzy lines, where transgressions might or might not trigger a response. War-fighting in the sense that nuclear weapons were no longer treated as an all-or-nothing “spasm”, to use Robert McNamara’s phrase, but were redesigned in all sorts of shapes and forms so that their actual use as a response to a wide range of threats was much more believable. The central insight of Schelling’s work is that a probabilistic promise to use nuclear weapons, a “threat that leaves something to chance”, combined with a range of nuclear options so that it’s easier to use them, creates a much more stable peace where nuclear weapons will never be used. Thanks in part to Schelling’s work, the US government got out of the business of making guarantees when it came to nuclear deterrence (as did the Russians, who matched our move to a war-fighting strategy pretty much step for step), and the world was far safer for it.
How, then, to explain the opposite tendency within domestic policy, where the government tends to embrace programmatic economic guarantees rather than push them away? Specifically, why have we moved from the war-fighting ambiguity of a Greenspan put to the MAD-like guarantee of open-ended QE? There is both a human impetus to transforming a probabilistic promise into a guarantee as well as a political impetus, and the Fed Chairmen and governors who determine monetary policy are much more prone to both than the generals who determine nuclear deterrence policy.
The human impetus is familiar to anyone who is constantly in the public eye, whether it’s a sports star or a Fed Chairman. There is such a public appetite for the dramatic story arc of a guarantee, and there is such an enormous public reward for “delivering” on that guarantee and becoming a Hero, that these public guarantees are made all the time in highly public endeavors even though everyone knows that the guarantor has no certainty of success, particularly in a team sport.
For every Joe Namath or Mark Messier you have at least two Patrick Ewings, who become Goats for their failed guarantees. Despite these odds, it doesn’t prevent new guarantees from being made every day. Stranger still, it doesn’t dissipate the public and media fascination with the act of making a guarantee, even though we all know it doesn’t make any sense. Such is the power of the Hero Narrative.
Generals don’t make guarantees – not because they are smarter than sports stars or Fed Chairmen – but because, unless we’re in the midst of a big war, no one knows who they are. I would bet that not one person in 100,000 can name the current Chairman of the Joint Chiefs of Staff, much less the head of the US Strategic Command. But everyone in the world knows who Ben Bernanke is, and that creates an enormous pressure to “step up” and be who everyone wants him to be: the Hero. We all know what a Hero does, including the person thrust into the role. He must inspire his team or his troops or his country, and that means making an unconditional guarantee of success. This is the story arc of the Hero, from Gilgamesh to Arthur to Washington to Bernanke. Like it or not, it comes with the job.
The political impetus to embrace programmatic guarantees is no less powerful than the personal. Governments protect the governed from threats. This is what they do. This is how they stay in power. I don’t mean that this is how specific individuals or political parties stay in power (although this may also be true). I mean that this is how a regime – the amalgamation of political norms and institutions that constitute what it means to be a government at any given point in time – maintains the consent of the governed. Regimes do not stay in power by succeeding wildly; they stay in power by avoiding abject failure and by responding publicly to perceived threats. Once embraced, programmatic governmental guarantees never just go away on their own. Not only do they become institutionalized and thus acquire bureaucratic inertia and support, but more importantly they become part of what it means to be American, or French, or Chinese, or what have you. The threats that any regime responds to may change over the decades, but the programmatic responses to those threats accrete and remain over time.
The perceived threat in the aftermath of the Great Recession is an inchoate fear of something that will make prices go down again. Worried that Congress might make some fiscal policy error? Worried about Obamacare? Worried about a Chinese hard landing? Worried that Europe might not get out of recession? Worried about the Middle East? Better keep QE going just to be safe. The future looms large today as a threat rather than as a promise, because the American regime (and by extension, the global regime) cannot withstand another nationwide decline in US home prices or a serious decline in the US stock market. Why not? Because the programmatic guarantees already made by the American regime (pensions, retirement insurance, medical insurance, poverty insurance, housing insurance, food insurance, banking insurance, etc.) cannot withstand a deflationary environment. It is politicallyuntenable for asset prices to go down again, and so they won’t. If that comes at the cost of massively pulling forward demand for risk assets, of creating the mother of all crowding-out effects in Treasuries, of creating a $4 trillion balance sheet to fund an umbrella guarantee program, of lowering the structural growth potential of the country and the world … well, so be it. The goal of any regime, any organism, is to maximize its chances of survival. Deflation is the perceived existential threat of our age, and this is the dragon our Heroes will guarantee to slay.
Providing a programmatic guarantee against asset price deflation does NOT mean that runaway inflation is just around the corner, any more than it means Social Security will go bust or that Obamacare will bankrupt the country. You can create a compelling story arc around any of these scenarios, to be sure, and maybe they will come to pass after all, but I wouldn’t ever want to bet against the ability of the American regime to maintain the status quo Narrative well enough to survive. As Milton wrote, you can get used to anything over time, even Hell, and from that perspective, programmatic QE doesn’t look half bad.
By the same token, however, providing a programmatic guarantee against asset price deflation does NOT mean that nothing has changed structurally in the relationship between State and Market. This time IS different, in the same way that the structural relationship between State and Market changed in the aftermath of the Long Depression of the 1870’s and the Great Depression of the 1930’s. There’s no sense in rending our clothes or gnashing our teeth over what’s happened … this is the business we’ve chosen, to quote Hyman Roth, and so we better get on with our business in as clear-eyed a fashion as possible. We are all suckers for a good story, but we can improve our decision-making under uncertainty by recognizing this innate human bias, by taking a broad historical view of current events, and by calling things by their proper names.