The Quest for Certainty

The Economics of Assumptions

By John Mauldin

“As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.”  – Albert Einstein

“To trace something unknown back to something known is alleviating, soothing, gratifying and gives moreover a feeling of power. Danger, disquiet, anxiety attend the unknown – the first instinct is to eliminate these distressing states. First principle: any explanation is better than none… The cause-creating drive is thus conditioned and excited by the feeling of fear …"   – Friedrich Nietzsche

“Very few beings really seek knowledge in this world. Mortal or immortal, few really ask. On the contrary, they try to wring from the unknown the answers they have already shaped in their own minds – justifications, confirmations, forms of consolation without which they can't go on. To really ask is to open the door to the whirlwind. The answer may annihilate the question and the questioner.”  – Anne Rice, The Vampire Lestat

The last two weeks we have been looking at the problems with models. First we touched on what I called the Economic Singularity. In physics a singularity is where the mathematical models no longer work. For example, models based on the physics of relativity no longer work if one gets too close to a black hole. If we think of too much debt as a black hole of sorts, we may understand why economic models no longer work. Last week, in “The Perils of Fiscal Cliff,” we looked at the use of fiscal multipliers by economists in order to argue for or against governmental economic policies. Do you argue for austerity, or against it? There is a model that will support your case, most likely using the same data that your adversary uses.

These letters have generated a great deal of positive response and conversation. While I very rarely suggest to readers to go back and read previous letters, but reading these may help you appreciate why it is so difficult to understand what is happening in the global economy today.

This week, in a somewhat shorter letter, we once again consider the vagaries of measurements and models. Growth of the US economy, we are told, was 2% last quarter. That number will of course be revised, but what is it we are measuring? Should we attach any importance to the measurement at all? The short answer to the last question is yes, but it is important to understand that there is no certainty in that number. Or at least not any certainty according to the generally accepted meaning of that word.

The Problem with Dynamic Condition-Dependent Multipliers

I use the above subhead with a great deal of irony. I garnered that phrase from a rather insightful letter from my friend Rob Arnott (founder of Research Affiliates, whose brilliant work is used to manage over $100 billion). I am going to start this week’s Thoughts from the Frontline with his letter, with my comments inserted in brackets [and italicized]. He was responding to last week’s letter on the problem with economic forecasts. Incidentally, Rob and his family spent five days in Italy with me this summer. There were many late-night discussions (and lunchs and dinners) about this topic.

“Most fiscal multiplier data is simplistic, trying to convert dynamic condition-dependent multipliers into static multipliers [i.e., if you adopt this tax cut or that spending increase you will have a very specific effect on the economy].  Why?  Because that makes economics, economic forecasting, and policy prescriptions pretty simple and easy to explain to reasonably intelligent people.  In so doing, ignore the impact of:

·       the starting levels of debt, magnitude of deficits, rate of change of demography

[This is important and at the root of Economic Singularity. If you start with massive debt, as the Greeks or Spanish have, stimulus spending has much less effect than if you have a very small debt. There is a significant difference between the economic policies of Japan and Brazil, for instance, because of the age of their populations. What works in one country may be counterproductive in the other.]

·       different multipliers between taxes and spending, and between increases or cuts in either [The academic literature suggests that there is a clear difference between the effects of income taxes and consumption taxes on the economy. Trying to use a one-size-fits-all multiplier on taxes or spending is bad economics. Much easier, of course.]

·    differential multipliers for temporary vs permanent tax changes (the mythical temporary tax cut and its famously useless multiplier)

·      different multipliers for categories of fiscal stimulus (bridge to nowhere – bad; repair creaking bridge between economic centers before they collapse – better)

·         the effects that employment policy choices can have on the fiscal multiplier (i.e., let’s not forget the role of the private sector!). 

“Move away from a static multiplier, and there are so many possibilities. 

“Most of the factors that affect multipliers are not easily controlled by government; and most changes inflict pain before they deliver their promised benefits.  Reciprocally, changes that deliver immediate benefit carry a daunting long-term price tag. 

“Consider the vicious cycle embedded in fiscal multipliers.  If aggregate tax rates are 50% (between income and VAT and property taxes, not to mention regional or city taxes) in most of Europe today, a multiplier of 2 creates infinite feedback [which brings us to a Point of Economic Singularity where the models will produce silly results, so economists just assume away any such condition]. 

“… in such a world, you could boost taxes by 1% of GDP and watch GDP drop 2%, producing no more tax revenue than before; rinse and repeat, until you’re beyond Greece.  At Hollande’s 75% rate (80% if the wealthy actually spend anything and incur VAT), you hit this cycle if the multiplier is 1.25.  And will the multiplier be affected by tax-rate levels?  Of course.  Higher taxes strangle the private sector so that the multiplier for incremental change in tax rates will increase. [Note: Rob is not arguing for the use of the specific multiplier, just pointing out that a single-variable multiplier can seriously distort the models.]

“The right answer is to not get into this mess to begin with.  After we’ve blundered into this situation, the predictable political reaction seems to be, ‘I can’t inflict pain on my watch.  Every pain must be immediately rewarded by gain.’  With a dearth of useful alternative ideas, this leads to the ‘kicking the can’ nonsense, which makes the eventual implosion far worse.

“The right answer, once we’re already on the gurney, needing glucose to stay alive, but needing to shed the consequences of past glucose overdose, is only a little more subtle.  First, we adopt the Hippocratic Oath: ‘First, do no harm.’  E.g., do nothing that carries lasting consequences that might exceed the near-term benefits.

“So, what are the possibilities? 

·      One thing is to try to understand dynamic condition-dependent multipliers, even roughly, over multiple time spans.  As the economics profession seeks simple models that can win Nobel Prizes, they’ve let us down by doing far too little of the work on condition-dependency or horizon-dependency. 

·        Absent convincing models on the multipliers, do not dismiss common sense!  Would borrowing and spending another $10 trillion produce a positive multiplier for next year?  Of course.  Over five years?  Of course not.  It’s no different from a family that boosts the GFP (Gross Family Product … i.e., family run-rate consumption) by buying a car they can’t afford.  Would stimulus diverted to pork projects boost near-term GDP more or less than long-term debt?  The latter is pretty obviously the correct answer.  It’s no different from a family deciding to buy bling rather than upgrading to a more reliable commuter car.  Common sense trumps mathematical models, every time.

·          Apportion austerity *and* stimulus, short-term pain and gain, in measures that deliver maximum deficit reduction, ideally without deep recession.  E.g., stimulus spending has an okay short-term multiplier, but it almost always delivers more long-term debt than near-term economic stimulus; spending cuts reverse this effect at a cost of near-term recession.  Tax hikes, if perceived as permanent (they usually are), have a terrible long-term multiplier; tax cuts reverse this … unless the bond market is spooked by default risk.  

“Liberalizing private sector labor laws is very underrated, because we have attached such a deeply ingrained [public sense of what is “right” and “fair”]:  [Labor laws] have a vast multiplier whenever regulation actively discourages new job creation, as is true throughout the developed world.  Get rid of the minimum wage?  How dare you?!?!  But, suppose it’s replaced with a negative income tax that allows flat rates at all income levels.  In effect, we’re replacing a regulatory minimum wage, imposed on employers and discouraging employment, with public funding of a minimum wage, supplanting much of the public welfare bureaucracy. [This is one of Rob’s more interesting proposals and would go a long way towards alleviating youth unemployment, a topic near and dear to this Dad’s heart. If you can hire a teenager for relatively little cost, you can give them a job and training that will mak e them more productive later in life.]

·          This means we can combine a high-multiplier positive policy choice, delivering immediate positive GDP gains, before (perhaps mere weeks before) instituting a lower-multiplier adverse policy choice, like cutting public spending, that has known long-term benefits at a cost of short-term pain.  The combination can offer goldilocks outcomes. [Or, what Rob might not argue, raise taxes while cutting spending in a compromise to control the deficit.]

“None of this happens without clear leadership at the top.  While my libertarian political leanings are well-known, there’s catnip in the sensible solution for both parties.  But, I doubt any of this will happen in time.  We ‘hire’ our political leadership for their perceived ability to respond well to crises, not to anticipate and avert crises.  Why would I say this?  Because averting crises usually requires short-term pain for long-term gain.  And no politician who deliberately imposes pain, successfully deflecting a crisis, will ever get credit for the success.  Why?  Because no one ever sees the crisis that didn’t happen!” 

It is here that I am more optimistic than Rob; and as my friend Newt Gingrich emphasized when he sat in on some of those late-night sessions in Tuscany, most politicians on all sides of this debate recognize the sheer magnitude of the disaster that it would be to avoid dealing with the deficit in what has now come to be the near term. Admittedly, they have different solutions, but they recognize the problem (note that I said most politicians – certainly not all). While Rob is right that no politician can run on a platform of cutting the things you like and raising taxes, if we’re not prepared to do something very much like that in the first part of 2013, the Fiscal Cliff we talk about will seem like merely stepping off a curb, compared to what comes next.

Measuring GDP

We will circle back to this discussion in a minute, but let’s visit this week’s announcement that GDP rose by 2% in the last quarter, up from 1.3% in the second quarter, which itself ended up being revised down almost 2% from the initial estimate. My guess is that we will see the same downward revisions over the next few months, as the other economic data last quarter was not robust. But clearly, we are not in a recession, just a Muddle Through Economy, as predicted here.

A 2% number is not bad, but there is more to it when you look at the underlying components. A large factor in the quarterly growth was defense spending, which leapt by a quite robust 13%. Personal consumption was up just 2%. Then there is inflation. If you think inflation was 2%, then the GDP number is overstated by 0.5%. Couple that with normal defense spending, and growth would have been less than 1%. That would not have been a political winner.

Inflation can be measured in several ways. GDP data does not use the Consumer Price Index, which shows inflation of more than 2%. You can get a much different GDP depending on what inflation number you use, and those numbers are dependent on what assumptions you make about how to figure inflation.

And while we all seem to use GDP, is that really the measure that makes the most sense in today’s world? Might we be better off targeting Gross Domestic Income, rather than looking at a consumption-based number like GDP? And isn’t Gross Private Production what we really need, rather than just an indicator that includes changes in government spending? At the end of the day, government spending can only be a function of what is produced in the private sector.

The Economics of Assumptions

We all want to have numbers that are “real.” But economics is different from accounting. Economics makes assumptions in almost all of the models it uses, and those assumptions come with biases. How many discussions do we get into that proceed along the lines of:

“Look at this statistic. It clearly goes up [or down] with GDP [or employment or…]. Therefore, if we could just fix ‘X,’ we would solve the world’s problems.”

For instance, I can clearly demonstrate to you that raising taxes on the rich will have no effect on their spending, if I use just one or two correlations in certain time frames. Throw in a few good stories, and the obvious conclusion is that we should raise taxes on the rich again and again. Just ask Monsieur Hollande – it’s their fair share.

Then I can just as easily show you that raising taxes on the rich will result in serious economic calamity. “Just see what it did in this situation. And see what cutting taxes did there.”

The counterargument then runs that your interpretation misses some other factor, so your conclusion is wrong. And so on and on. This goes back to the quote from Anne Rice at the beginning of the letter. While her character was talking about another form of knowledge, the observation applies doubly to economics. Here it is again:

“Very few beings really seek knowledge in this world. Mortal or immortal, few really ask. On the contrary, they try to wring from the unknown the answers they have already shaped in their own minds – justifications, confirmations, forms of consolation without which they can't go on. To really ask is to open the door to the whirlwind. The answer may annihilate the question and the questioner.”

Human beings seek certainty. We actually get an endorphin rush when we get an explanation for something we do not understand. Whether it’s religion, politics, philosophy, a crossword puzzle, or economics, we want to be able to come to a definite conclusion that we think is correct. There is psychological rest in certainty, along with the physiological rewards). Models, even flawed ones, give us the illusion of certainty. We need to be careful of what illusions we cling to.

Economics becomes quite a problematic discipline when it tries to create mathematical models that are supposed to guide political philosophy and praxis. So many assumptions have to be made to get to a result, that basing policy on a simplistic model is dangerous.

One size does not fit all, and past performance really does not indicate future results. The entire economic environment must be taken into consideration. We cannot extrapolate simplistically from the Reagan or Clinton years and say, “If we just reverted to those policies, we could get the same results.” Only if you could change all the other variables that are beyond the control of the government!

Models can be useful, but they are not exact. They give us a sense of direction. Using them is more like navigating by the North Star than using a GPS system. The more variables that enter into the actual situation, the less likely we are to be able to come up with that one “easy-button” policy prescription.

In the end, the only real tool we are left with is common sense, guided by our models and an appreciation of history. We “know” that, in general, the lower the price the higher the demand. If you tax something, you will get less of it.

We get that we can’t let financial institutions run amok. There have to be some protections for the public. Debt is useful until it becomes a burden, and we have to be careful in how we use it. We can come up with dozens of such truisms, based on common-sense wisdom.

We elect politicians and then expect that somehow the world will improve in accordance with their promises. What we really need to do is try to see what general direction they are leading us in and base our votes and our personal decisions on whether we like that direction. But to trust an economist, or even worse a politician, with a model? That can be dangerous.

Let’s close with a quote from my favorite central banker, Richard Fisher, who is president of the Dallas Federal Reserve.

“It will come as no surprise to those who know me that I did not argue in favor of additional monetary accommodation during our meetings last week. I have repeatedly made it clear, in internal FOMC deliberations and in public speeches, that I believe that with each program we undertake to venture further in that direction, we are sailing deeper into uncharted waters. We are blessed at the Fed with sophisticated econometric models and superb analysts. We can easily conjure up plausible theories as to what we will do when it comes to our next tack or eventually reversing course. The truth, however, is that nobody on the committee, nor on our staffs at the Board of Governors and the 12 Banks, really knows what is holding back the economy. Nobody really knows what will work to get the economy back on course. And nobody – in fact, no central bank anywhere on the planet – has the experience of successfully navigating a return home from the place in which w e now find ourselves. No central bank – not, at least, the Federal Reserve – has ever been on this cruise before.”

We indeed have not been on this cruise before as a nation and as a world. We know what happens when one country or another runs up against the limits of borrowing power. But when the bulk of the developed world does? Another cruise, indeed.