The Great Debate – The Real Story?

The Great Debate© is presented by to expand our understanding of various topics of interest. In this Great Debate©, the topic is whether the government should introduce a second stimulus. In a NYT Op-ed article, Nobel Laureate Professor Paul Krugman argues the first stimulus did not solve the economic ills because it was too small. Steven Hansen counters.   We present Professor Krugman’s opinion in its entirety, then the response of Steven Hansen.

The Real Story


Next week, President Obama is scheduled to propose new measures to boost the economy. I hope they’re bold and substantive, since the Republicans will oppose him regardless — if he came out for motherhood, the G.O.P. would declare motherhood un-American. So he should put them on the spot for standing in the way of real action.

But let’s put politics aside and talk about what we’ve actually learned about economic policy over the past 20 months.

When Mr. Obama first proposed $800 billion in fiscal stimulus, there were two groups of critics. Both argued that unemployment would stay high — but for very different reasons.

One group — the group that got almost all the attention — declared that the stimulus was much too large, and would lead to disaster. If you were, say, reading The Wall Street Journal’s opinion pages in early 2009, you would have been repeatedly informed that the Obama plan would lead to skyrocketing interest rates and soaring inflation.

The other group, which included yours truly, warned that the plan was much too small given the economic forecasts then available. As I pointed out in February 2009, the Congressional Budget Office was predicting a $2.9 trillion hole in the economy over the next two years; an $800 billion program, partly consisting of tax cuts that would have happened anyway, just wasn’t up to the task of filling that hole.

Critics in the second camp were particularly worried about what would happen this year, since the stimulus would have its maximum effect on growth in late 2009 then gradually fade out. Last year, many of us were already warning that the economy might stall in the second half of 2010.

So what actually happened? The administration’s optimistic forecast was wrong, but which group of pessimists was right about the reasons for that error?

Start with interest rates. Those who said the stimulus was too big predicted sharply rising rates. When rates rose in early 2009, The Wall Street Journal published an editorial titled “The Bond Vigilantes: The disciplinarians of U.S. policy makers return.” The editorial declared that it was all about fear of deficits, and concluded, “When in doubt, bet on the markets.”

But those who said the stimulus was too small argued that temporary deficits weren’t a problem as long as the economy remained depressed; we were awash in savings with nowhere to go. Interest rates, we said, would fluctuate with optimism or pessimism about future growth, not with government borrowing.

When in doubt, bet on the markets. The 10-year bond rate was over 3.7 percent when The Journal published that editorial; it’s under 2.7 percent now.

What about inflation? Amid the inflation hysteria of early 2009, the inadequate-stimulus critics pointed out that inflation always falls during sustained periods of high unemployment, and that this time should be no different. Sure enough, key measures of inflation have fallen from more than 2 percent before the economic crisis to 1 percent or less now, and Japanese-style deflation is looking like a real possibility.

Meanwhile, the timing of recent economic growth strongly supports the notion that stimulus does, indeed, boost the economy: growth accelerated last year, as the stimulus reached its predicted peak impact, but has fallen off — just as some of us feared — as the stimulus has faded.

Oh, and don’t tell me that Germany proves that austerity, not stimulus, is the way to go. Germany actually did quite a lot of stimulus — the austerity is all in the future. Also, it never had a housing bubble that burst. And with all that, German G.D.P. is still further below its precrisis peak than American G.D.P. True, Germany has done better in terms of employment — but that’s because strong unions and government policy have prevented American-style mass layoffs.

The actual lessons of 2009-2010, then, are that scare stories about stimulus are wrong, and that stimulus works when it is applied. But it wasn’t applied on a sufficient scale. And we need another round.

I know that getting that round is unlikely: Republicans and conservative Democrats won’t stand for it. And if, as expected, the G.O.P. wins big in November, this will be widely regarded as a vindication of the anti-stimulus position. Mr. Obama, we’ll be told, moved too far to the left, and his Keynesian economic doctrine was proved wrong.

But politics determines who has the power, not who has the truth. The economic theory behind the Obama stimulus has passed the test of recent events with flying colors; unfortunately, Mr. Obama, for whatever reason — yes, I’m aware that there were political constraints — initially offered a plan that was much too cautious given the scale of the economy’s problems.

So, as I said, here’s hoping that Mr. Obama goes big next week. If he does, he’ll have the facts on his side.

Facts and Opinions

by Steven Hansen

I am not an economist by education – but an Industrial Engineer. Most engineering disciplines are hard sciences where theories must be proven through controlled experiments – and then confirmed through application of that theory. When contradictions occur during application, studies are undertaken to identify the reasons.

You will not read of opposing proven theories in engineering. The theories are tested and applied through literally millions of cycles. A theory works or it does not. Proven engineering theories meet the test of being fact.

There is no such thing as a “proven” economic theory. There has been no controlled experiments on any theory as you cannot identify and control all of the dynamics. The closest you can get to proof is that a theory “seems” to work.

And any proof must be confirmed by repeating experiments hundreds or thousands of times. Economists “prove” theories by studying one or more (usually much less than 10) events. The dynamics surrounding most events are significantly different.

So what are the facts?

1) We stimulated the economy much more than $800 billion. You must throw in TARP, Fed intervention of $2 trillion, plus a whole series of piecemeal legislation. The total intervention was significantly more than $2.9 trillion hole the CBO said we had in our economy.

2) What is the correlation between the size of economic shortfall and stimulus? Stimulus by definition is a boost – not a replacement for – economic activity.

3) Did all or most of those who said the stimulus was too big predicted sharply rising rates? I know some did. Could it be true that if the stimulus was “large enough” there would have been rising rates? This is standard economic debate – say what you want because no one can prove you wrong.

4) The economic dynamics in Germany are too different than the USA. Comparisons are inoperative.

So, as I said, here’s hoping that Mr. Obama goes big next week. If he does, he’ll have the facts on his side.

There are no facts to support a position that additional stimulus will work, nor facts to support that it would not work. The “facts” here are opinions – and everyone has one..

5 replies on “The Great Debate – The Real Story?”

  1. A couple of points if I may,

    1) the amount of “spend” that got caught up in excess reserves in the banking system can’t be counted as stimulus.

    2) the semantics of boost versus replacement is not relevant. The money either went into the GDP figures at least once or it didn’t.

    3) some of this is purely mathematical

    4) there is lots of room for discussion of the quality of the spend in terms of whether the goods/services produced were highly necessary, were going to be spent anyway so it was just a timing difference and would contribute to future economic efficiency or not and if not, how much of the spend could have got greater long term benefits. Offsetting this is some of the argument about the urgency of the spend and in this I find Eichengreen and O’Rourke’s analysis at Voxeu on the size and speed of the downturn in trade and economic activity compared to the great depression very useful. Some tradeoff of quality of spend for speed was, in my opinion, warranted. Havng said that, there was clearly some very poor quality spending, maybe as high as 15% is my unsubstantiated guess. But even the inefficient spend went somewhere. The bloated margins went to company profits which then sustained shareprices and was partially recycled as dividends, the bloated payrolls got used to repay debt or keep buying goods/services which kept people employed. This is not to say that better quality spend would not have been better and that different targets for the spend may have had more chance of being respent within US rather than on imports

    5) the basic economic identity requires that private sector change has to equal the total of a) public sector change and b) external sector change, so big deficit spend must by definition increase GDP over what it otherwise would have been unless it leaks to the external sector.

    But then, my formal economics study is 35 years ago and most of it was disproven by the failure of most economists/all classical economists to see the crashes that we have had.

  2. Paul, appreciate your additional points.

    I was careful in crafting my response to Professor Krugman. My theme is simply you cannot prove any point on stimulus. Pick a point, and I can counter. No one can prove anything.

    Therefore, the only argument I have with your comment:
    “the semantics of boost versus replacement is not relevant. The money either went into the GDP figures at least once or it didn’t”.

    Stimulus does not have to go into GDP. Existing real estate sales is NOT in GDP – yet this effects other elements of GDP. Money that went into bank balance sheets are not in GDP – yet being able to be in a position to lend money for GDP activities is. The Fed buying treasuries did not go to GDP, but freed up other money that could.

    Stimulus should not a replacement for GDP but targeted factors to cause the economy to accelerate. If you will not spend the effort to target, then you use a shotgun approach which much of the money is simply “replacement for GDP”. Government spending has a significantly less GDP multiplier than private spending

  3. Steven, you write:

    “There is no such thing as a “proven” economic theory. There has been no controlled experiments on any theory as you cannot identify and control all of the dynamics. The closest you can get to proof is that a theory “seems” to work.”

    First, just as a point of clarification we like to say, “mathematicians and philosphers prove, scientist gather evidence”

    However, your basic point of the lack of repeat controlled experiment holds true for any observational science. There are no repeated trials on the Big Bang, the formation of comets, the evolution of species, the movement of tectonic plates or the extinction of the dinosaurs. One can repeat the same exact weather or oceanographic study under the exact same conditions. Indeed, not to wax to philosophical but control is a matter of degrees. The only things which are exactly alike are the fundamental and their small scale constructions. It might even be reasonable to suggest that two molecules of carbon dioxide are exactly alike. Its much less likely that two strands of DNA are exactly alike down to every single base pair.

    Up until the last 70 years or so it would have been hard to think of controlled experiments on the structure of molecules. At least at a level of control greater than we have over the economic policy. One could not hold a molecule still while it was poked and prodded, confident that the only thing affecting it was the measuring instrument.

    Yet, we accumulated vast arrays of information about these areas. Much of it could even be called deduced fact. This is the same with economics.

  4. Karl, I have read your work and appreciate your comment. It will take a lot of convincing for me to accept that economics rises to the same level as any “hard” science such as physics or biology. more samples, real hard data to support conclusions, less diversity in opinion because of preponderance of evidence.

  5. Karl and Steve – – –

    The nature of proof is not well understood by many but I think you both clearly understand the subtleties. Some reading the two comments may be confused, so let me weigh in to confuse them some more.

    In science, proof is based on the absence of contradictory evidence. A scientist proposes a hypothesis. This can be tested with a combination of observations and mathematical proof. If the hypothesis is substantiated through repeated test, it becomes a theory. A theory that passes the “test of time” becomes a law.
    The mathematical proofs of theories and laws will necessarily rest on a set of assumptions, often that prior theories and laws are true. That may not be an irrefutable foundation. For example Newton’s Laws of Motion became insufficient for microscopic and sub-microscopic phenomena when it became apparent that probabilistic descriptions were needed to explain such behaviors. The old precisely fixed time and space relationships of Newton simply didn’t work in that regime.

    All of a sudden, in the early 20th century, the dual nature of light (wave and particle characteristics) evolved to explain behaviors at the atomic and sub-atomic level. What had been a beautifully structured classical physics description of the world became a fuzzy and uncertain menu of wave equations and statistical mechanics. Heisenberg postulated (now widely accepted but rejected when proposed) that we can not precisely know both the position and energy of anything at the same time.

    As confusing as all this transformation seems, it is still quite orderly when compared to the field of economics. The state of affairs here is far behind the rigorous structure of modern physics and the probabilistic theories and laws that have evolved. Economists do not deal with irrefutable laws for large scale objects that can be or have been extended into probabilistic theories and laws to explain micro behavior.

    Instead, economists deal with correlations of pairs of parameters. Perhaps employment correlates well with home sales, for example. However, a complex statistical hypothesis of home sales as a function of employment, interest rates, birth rates, average household size, bank capitalization ratios, etc. is the sort of thing that is rarely attempted. And for good reason because, it is likely that, no matter how long the dependent variable list becomes, the list of dependent variables is likely to be incomplete and the number of independent observations possible is probably going to be less than the number of unknowns to be solved for. (Think solving simultaneous equations.)

    It is true that the science of physics is still incomplete. After all, there still is no “theory of everything”. There may never be. However, there are theories and laws that explain vast areas of observation and have withstood decades and centuries of scrutiny. Economics has not come close to that level of description and definition.

Comments are closed.