A mental model for understanding policy errors and consistency anchors using the Fed and Greece

One of the great things about markets and economics is that there is a lot of uncertainty built into them because, unlike traditional macroeconomic modelling assumptions, real people are unpredictable and irrational. This makes life uncertain, and at economic and market extremes it introduces Knightian uncertainty that just can’t be modelled with mathematical models. We are in one of these times in which Knightian uncertainly is everywhere. In that vein, I am going to put down a few thoughts on how behavioral economics might inform us here.

Nothing I write today should be taken as anything more than random musings. I don’t have a formal model. I am just trying to tease out a mental model of how to think about uncertainty in a world in which geopolitical risk is great and policy error looms at every turn. What made me think about writing this piece was re-reading my recent pieces on the Greek negotiation. I believe there is a deal to be had. But the stakes are large and the negotiating positions polarized. Even more importantly, ideas about what the other guys best alternative to a negotiated agreement are very different from one’s own idea. Many of the Troika policy makers think they have firewalls in place which will limit contagion from a negative outcome in Greece. They have the whole alphabet soup arsenal ready to spray sloppy, soupy letters of ease on the remaining periphery countries – OMT, QE, ESM. And market rates of interest do suggest Greece is a special case in many ways.

Yet, on the other hand, the Greeks are singing a different tune. Greek finance Minister Yanis Varoufakis says “the euro is fragile. It is like a house of cards. If you pull away the Greek card, they all come down.” He seems to think Europe is still weak. And some of the data support that contention.

For example, the political climate in the periphery is becoming radicalized by the years’ long struggle against austerity and anti-austerity political groups are gaining support. In Spain, for example, Podemos is actually leading the polls while the two main parties collectively are polling under 40%. That’s breathtaking given Podemos didn’t exist before the sovereign debt crisis.

Moreover, the economic conditions in the periphery really aren’t that great. In Italy, where nominal GDP growth had been weak, even before the global financial crisis, we are looking at an economy with 133% government debt to GDP and no real lender of last resort. It is also is a nation laden with banking problems. German newspaper Die Welt recently highlighted the fact that Italian banks have 333 billion euros of problem loans on their books (link in German). The Italian banks also had by far the worst report card under the stress tests conducted during the ECB’s asset quality review.

This fact is highlighted by the massive loss reported by Monte dei Paschi today. And Monte dei Paschi is the only bank that received any bailouts in Italy and they only received 4 billion in guarantees, not actual government capital, since Italy’s existing government debt level makes that extremely unpalatable. According to Eurostat, Italy’s 4 billion euros of bank bailouts stands in stark contrast to Germany’s 250 billion, Spain’s 60 billion, Ireland’s 50 billion, Greece’s 40 billion and Portugal’s 18 billion. And Portugal, despite the 18 billion euros in bailout money is in a precarious position with total debts 380% of GDP, a budget defict still over 5% and thus vulnerable to more austerity. We are not out of the woods here by a long stretch.

Thus, the potential for “policy error” in my view is mostly about the Troika thinking the Greeks have no alternative to a negotiated agreement and pressing this to the brink such that no deal results and Greece defaults.

So let’s think about how and why we got here and I’ll briefly use the same logic regarding the Fed as another example of how this mental model could be applied.

Almost four years ago, I wrote a post about our innate need for consistency. I framed it as follows: “Human beings are hardwired to reduce complexity and to deal with uncertainty in order to be able to make quick decisions. It is amazing how well we navigate the complexity not just our physical environment but of our social environment as well by employing a number of tactics automatically called heuristics. I have talked about some of these tricks like confirmation bias or confabulation in the past. So let me weave this into a story of how pundits and politicians strive for consistency and how they unwittingly constrain themselves in order to achieve that goal. I think this is an important facet of the political economy we are now seeing in the sovereign debt crisis in Europe and in economic policy in the United States.”

The point is this: most people are anchored by a prior decision that they will defend tooth and nail, irrespective of whether doing so is actually advantageous. The reason for this is the need for consistency.I suggest you go ahead and read that post on consistency because what I am saying now follows from it (click link here). But the synopsis in terms of making a mental model is that consistency is a defining trait for what we perceive as honest, trustworthy, rational human beings. People who are inconsistent are seen as illogical, deceitful and untrustworthy. And so we strive to be as consistent as we possibly can – even when doing so is actually not in our best interests.

Moreover, consistency offers people a “safe ideological hiding place” from a complex world. Why challenge existing decisions if they were made based on sound thinking that fits your rational ideological view? That would only invite scrutiny and subject you to taunts of being “inconsistent”.

What all of this means is that human beings sometimes make decisions that are globally rational in that they maximize ideological consistency across time but that are locally suboptimal, creating Knightian uncertainty that generates market tail risk. The Greek negotiations are a perfect example of where this can happen. The Germans could have boxed themselves into a corner ideologically that is hard to wriggle out of without looking like they ‘blinked’ or ‘caved’ or ‘capitulated’. These are just euphemisms for changed positions in a negotiation. And they are all negative euphemisms because they reinforce the connections human beings make between valour and consistency, weakness and inconsistency. And the fact that we have the alphabet soup to splatter across the periphery makes it all the more reasonable to stay the course and risk blowing up the eurozone.

At the same time, Syriza was elected literally weeks ago to resist austerity and roll back the Troika programs. How can Tsipras and Varoufakis sell their deal to the Greek people if they are seen as weak and having backed down? To be perceived by the Greek electorate as trustworthy and honest – that is to be perceived as consistent – no deal can be reached under any circumstances, unless it rolls back austerity and ends the Troika’s strangle hold on Greek economic life. To bring back anything less than a roll back of the Troika programs is to have ‘capitulated’ or ‘sold out’.

This is how policy errors happen.

And when policy errors occur, Knightian uncertainty rears its ugly head and that’s when the black swan events occur. This is why everyone is saying we could be on the verge of another ‘Lehman’. That was the last time a major policy error produced market tail risk. And it was catastrophic.

The same kind of policy error could be at hand with the Fed. Why? The Fed has produced a panoply of data points, literally, that chart out its projections of where interest rates should be going forward. And having done so publicly in order to provide the markets with better ‘forward guidance’, the Fed may have boxed itself into a corner on rate hikes. I know Fed Chair Yellen says that the Fed’s decisions are data dependent. But are they really? I mean how data dependent? My sense is that the Fed has a tightening bias. It has backed up that tightening bias with its famous dots chart showing where Fed officials want rates to be. And all of this is pointing to rate hikes mid-year. The incoming data will have to be pretty awful for the Fed to climb down and back away from its timetable. Otherwise, it will look inconsistent, lose credibility, invite front-running, and destroy the impact that forward guidance has.

I’d like to think the U.S. is on a sustainable upward economic trajectory. But I am not at all certain. In fact, to the degree you believe economic data are mean reverting, you should be concerned that we are at a point nearer the end of a cycle, where a downshift into recession can be nudged along by monetary policy.

Take a look at this chart from Albert Edwards on S&P500 earnings growth:

What I see is a downshift in profitability. Call it the strong dollar, increased labor costs, mean reversion after financially engineered growth, whatever. The numbers are still pointing down. And in a world in which both business spending and inventory accumulation are surprising to the downside, a nudge from the Fed could be extremely negative.

Again, I would like to believe we are beyond this point. Economic data are pointing up. But the Fed could be hiking rates just as the economy is pausing. We just don’t know. Nor do we know how the Fed will react to incoming data. There is a ton of uncertainty out there right now. And that means the risk of policy error is high.

I will leave you with the quote from Frank Knight that makes the most sense here. This comes from his book on Risk, Uncertainty, and Profit, which defines what is now known as Knightian uncertainty:

“Uncertainty must be taken in a sense radically distinct from the familiar notion of Risk, from which it has never been properly separated…. The essential fact is that ‘risk’ means in some cases a quantity susceptible of measurement, while at other times it is something distinctly not of this character; and there are far-reaching and crucial differences in the bearings of the phenomena depending on which of the two is really present and operating.”