The USA could benefit from overhauling the constitution and government. The current system reflects the highly stratified class society of the early industrial revolution period in Great Britain.
2slugbaits
JDH The debt ceiling provision is (historically speaking) a fairly recent addition to the budget & execution process. For most of US history the Congressional authorization was all that was needed. If there comes a time when the Treasury is not able to meet creditors’ demands, my best guess is that the Treasury would ignore the debt ceiling and just continue to pay out. This would set up something of a constitutional “crisis,” but the debt ceiling provision is constitutionally dubious anyway.
I’m not sure I’m following your point about 2 Aug not being a hard date. If you’re saying that we don’t really know whether it might be 3 Aug or 1 Aug, then okay, I’ll grant you that. There’s always going to be some error around the estimate. But if you’re suggesting that the Treasury could really push things out weeks beyond 2 Aug, then I find that claim a little hard to swallow. Somebody is not going to get paid. [Interestingly, SS benefit checks are scheduled to go out on 3 Aug.] It would be a very bad negotiating tactic. If you threaten something, then you better follow through; and since executing the threat is something that the Administration does not want to do, the way you make that threat credible is by not giving yourself any alternative. So I think 2 Aug is the Treasury’s best estimate of a hard date, although they may have just enough wiggle room to make the 3 Aug SS benefit date.
But hard date or soft date, what really counts is what the markets think. If we’re in the 3rd week in July and the parties are no closer to agreement than they are now, then I don’t expect the whiz kids at Goldman-Sachs to just sit on their butts waiting to see if 2 Aug is really a hard date or not. The interesting complication is that late July may also be when the EU has to start dealing with some hard dates of their own regarding Europe’s periphery. What’s an investor to do?
Mary
Good Interview
ppcm
Thanks,very precise and clear as always.The decor is medieval not matching the expected flamboyant gothic of our golden era.
Ricardo
Professor,
Great interview. It is hard to explain complex subjects in a couple of minutes and even then the talking head talks more than the person who actually has the expertise.
There are a minority of us who make the same point you did about a disconnect between spending, taxes, and debt. We recognize that the congress spends first and looks for a way to pay later whether taxes or debt. But the question is what are the parties doing to reconnect the three? What are you hearing from the Democrats? They want to ignore the impact of taxes and the out of control debt and not only keep spending high but even increase spending.
It seems that the Republicans actually are attempting to make the connection, though weakly, in their connecting spending cuts to increasing the debt ceiling. It is the Democrats who what to keep the two separate.
On taxes Christina and David Romer have given us the answer. A 1% increase in taxes means a 3% decline in GDP. Increasing taxes is a non-starter if you actually believe our problem is a lack of growth. They have also pointed out that a 1% decrease in taxes translates into a 2-3% increase in GDP.
If the deficit is going to increase due to spending or to tax cuts doesn’t it make a lot more sense to decrease taxes and get a 3% increase in GDP rather than maintaining high spending and raising taxes causing a 2-3% decrease in our anemic 1.8% growth. Doesn’t that mean that taxing will actually push us into GDP decline?
The plan is actually very simple. Cut taxes to gain that 3% growth, cut spending to return to sanity concerning the deficit, lock the debt limit to no additional increases so that 1 year from now we are not back dealing with this issue like we did in January 2010.
Mark A. Sadowski
Ricardo,
It is my understanding, based on things she and David have said, that Christina Romer believes that the tax multiplier is no larger than the spending multiplier. And she certainly does not believe that a balanced-budget equivalent reduction in taxes and spending provides any fiscal stimulus at all.
To compare the fiscal multipliers estimated previously with the Romers’ estimate would be inappropriate. They used a very different (and in some ways revolutionary) methodology. Thus it is not comparable. For example, previous studies have not fully controlled for the tax increases that often accompany spending increases. Thus they are very likely to understate the effects of spending increases alone. Valerie Ramey’s study in particular assessed the impact of the Korean-War military spending increase without taking account of the fact that it was accompanied by a large tax increase.
What the Romers’ study showed is not that tax cuts are uniquely effective, but rather that failing to consider the reasons for tax policy changes leads to an underestimation of their effects. Because these issues of omitted variable bias are likely to be as strong for spending as for tax changes, the most reasonable interpretation of her paper is that all types of fiscal stimulus are more potent than conventional estimates would lead us to believe.
A passage of the paper that helps to drive the correct interpretation of the paper home is the following:
“Following a spending-driven tax increase, real GDP on average rises moderately, reaching a maximum of 0.7 percent after two quarters (t = 1.5). Thereafter the effect fluctuates irregularly around zero and is always far from significant. Thus looking at how the economy behaves after tax changes driven by spending changes yields estimates of the effects of tax changes that are starkly different from those based on exogenous tax changes.”
In short tax increases coupled with spending increases yielded no statistically significant effect on the economy, and in fact, in direct contradiction to unrealistic supply side enthusiasts’ beliefs, on average, at least temporarily, they increased GDP. (I have no problem with the idea that taxes have a long run effect on AS, but we are of course talking about the short run effect on AD here.)
In the final analysis, the Romers’ paper found that, with the exception of countercyclical tax changes (possibly because there were too few, or because the economic downturn obscured their effect), the only type of tax change that had a statistically significant effect on GDP were those that changed the federal budget balance. This is hardly a result that supports your plan of action.
2slugbaits
Mark A. Sadowski Thanks for doing the heavy lifting. It saved me a lot of effort. Ricardo has brought up the Romer & Romer paper several times, and Menzie has politely suggested to Ricardo that he might want to reread their paper (assuming he has actually read it in the first place). As you pointed out, the Romer & Romer paper actually undercuts Ricardo’s argument. And like a lot of folks who post here, Ricardo doesn’t understand the difference between an AD curve and an AS curve. Today all of the action (or inaction) is on the AD curve. Ricardo still thinks it’s 1979.
Jim A
ISTM that in the face of Congressional guidance to spend X, Tax Y, and borrow no more than Z, the only legal option left to the administration is to sell assets. There’s all that yellow stuff in Kentucky, that might allow the US to operate a few months more…
Anonymous
Romer paper is at http://elsa.berkeley.edu/~cromer/RomerDraft307.pdf
In Section VI Conclusions is the following excerpt:
“In terms of consequences, there are six main findings. First, tax changes have very large effects on output. Our baseline specification suggests that an exogenous tax increase of one percent of GDP lowers real GDP by roughly three percent. Our many robustness checks for the most part point to a slightly smaller decline, but one that is still well over two percent. Second, these estimated effects are substantially larger than those obtained using broader measures of tax changes, such as the change in cyclically adjusted revenues or all legislated tax changes. This suggests that failing to account for the
reasons for tax changes can lead to substantially biased estimates of the macroeconomic effects of fiscal actions. Third, investment falls sharply in response to exogenous tax increases. Indeed, the strong response of investment helps to explain why the output consequences of tax changes are so large. Fourth, the output effects of tax changes are highly persistent. The behavior of inflation and unemployment
suggests that this persistence reflects long-lasting departures of output from its flexible-price level, not large effects of tax changes on the flexible-price level of output.”
Sounds like Ricardo quoted the paper very accurately.
Ricardo
Mark and Slug,
What the Great Depression, the Great Inflation, and the current Great Recession have proved and are proving for the umpteenth time is that the so-called spending multipliers don’t work at best and probably create bubbles to make things worse. Now that we have followed Christina and David’s spending multiplier idea to abject failure why not try the tax cut multiplier and see if it works again:
as it did in the UK after the Napoleonic Wars,
as it did in the US and Italy after WWI,
as it did in Germany and the US after WWII,
as it did with the Kennedy-Johnson tax cuts in the 1960s,
as it did with the Reagan tax cuts,
as it did with Clinton’s capital gains tax cuts,
Just to mention a few.
Mark A. Sadowski
Anonymous,
The issue is not the estimated effect of an exogenous tax change. The issue is whether the Romers’ research implies that tax multipliers have a different effect from spending multipliers. I think it’s quite clear from their results concerning endogenous tax changes that this is not the case. In the final analysis the tax change Ricardo is proposing would not even be classified as exogenous by the Romers.
Ricardo,
Five of the eight historical incidents you list were examined by the Romers’ paper. And in most cases your anecdotal evidence involves simultaneous spending changes and is conditional on the monetary policy response function.
Rich Berger
From the Abstract to the Romers’ article:
“This paper investigates the impact of changes in the level of taxation on economic activity. The paper uses the narrative record presidential speeches, executive-branch documents, and Congressional reports to identify the size, timing, and principal motivation for all major postwar tax policy actions. This narrative analysis allows us to separate revenue changes resulting from legislation from changes occurring for other reasons. It also allows us to further separate legislated changes into those taken for
reasons related to prospective economic conditions, such as countercyclical actions and tax changes tied to changes in government spending, and those taken for more exogenous reasons, such as to reduce an inherited budget deficit or to promote long-run growth. We then examine the behavior of output following these more exogenous legislated changes. The resulting estimates indicate that tax increases are highly contractionary. The effects are strongly significant, highly robust, and much larger than those obtained using broader measures of tax changes. The large effect stems in considerable part from a powerful negative effect of tax increases on investment. We also find that legislated tax increases designed to reduce a persistent budget deficit appear to have much smaller output costs than other tax
increases.”
They distinguish between endogenous and exogenous tax increases. As I understand the distinction, endogenous tax increases are those that follow through the natural working of the tax system (e.g., increasing incomes generate higher taxes) versus exogenous increases which are the result of deliberate legislation. They seem to say very unambiguously that the exogenous tax increases (such as those embodied in CBO’s extended baseline and alternative scenarios) are contractionary. The best that can be said is that they are less contractionary when devoted to reduction of persistent deficits. I think their paper provides additional ammunition in favor of the Ryan plan which does not require additional tax increases.
Incidentally, in a search of this site, I found a post back in 2007 where Prof Hamilton commented on this study and which was very helpful to me in understanding its basic thrust.
Jeff
In the final analysis the tax change Ricardo is proposing would not even be classified as exogenous by the Romers.
Mark,
Once the effect of an explanatory variable has been determined, that effect does not change simply because a new observation is determined to be endogenous.
Ricardo
Mark wrote: Ricardo, Five of the eight historical incidents you list were examined by the Romers’ paper. And in most cases your anecdotal evidence involves simultaneous spending changes and is conditional on the monetary policy response function.
Mark,
True, but the Romers do not say that tax decreases were not part of the growth. What they said is that there are other contributing factors. So, once again I ask, why not try it? At the worst it seems that the Romers are saying it may have little to no effect, but they are not denying that it could have a great effect.
When we know one thing has failed but another has worked, why resist the one that works?
Mark A. Sadowski
Rich Berger,
You wrote:
“As I understand the distinction, endogenous tax increases are those that follow through the natural working of the tax system (e.g., increasing incomes generate higher taxes) versus exogenous increases which are the result of deliberate legislation.”
Your understanding is incorrect. Both endogenous and exogenous tax changes are in fact the result of deliberate legislation. If you read the paper you’ll see they spent a great deal of time going through the legislative record classifying tax changes as one or the other. They state quite clearly that they specifically wanted to avoid changes in revenue that were purely the result of changes in income.
Endogenous tax changes include both countercyclical tax changes and tax changes coupled to spending changes (so as to minimize the change in budget balance). Exogenous tax changes are essentially all other tax changes.
Jeff,
You wrote:
“Once the effect of an explanatory variable has been determined, that effect does not change simply because a new observation is determined to be endogenous.”
The Romer’s motive was to isolate tax changes that changed a budget balance but which were not in response to economic conditions. Their concern was that the effect of tax changes taken in response to economic conditions might have their effect obscured by those same conditions.
Thus the econometric endogeneity in this context refers to whether or not it was in response to economic conditions. The fact that tax changes coupled with spending changes were classified as “endogenous” was merely to separate them from noncountercyclical tax changes that changed a budget balance.
The key to understanding this paper is too keep in mind it was done in the context of research on macroeconomic stabilization policy (short run growth). They were interested in detecting more clearly the effect of a change in the budget balance on aggregate expenditures by isolating it from economic conditions that might have obscured that effect. The rationale for working with legislated tax changes rather than spending changes seems to be that the data was easier to work with, although in theory the same methods could be applied to spending changes. In short it is part of the larger body of New Keynesian literature.
There is a small body of literature on the effects of tax level and tax structure on long run growth but this is not part of it. (I’m intimately familiar with it, as it forms the literature review to my dissertation.) That is part of the larger body of Growth literature.
Those trying to read supply side results into a paper that is demand side in focus are quite simply barking up the wrong tree.
Mark A. Sadowski
Ricardo,
To the extent that monetary policy accomodates fiscal policy, I believe fiscal policy is effective in regulating aggregate demand. Between QE1 and QE2 the Fed did very little one way or another so I have no reason to doubt that the effects of the discretionary fiscal stimulus have been what the major private and public forecasters have estimated.
So in short I do not believe the discretionary fiscal stimulus has been ineffective. It’s just that we have a very serious recession which requires stronger action. And I certainly do not believe that the tax multiplier is any larger than the spending multiplier.
In any case, as I do not believe in the liquidity trap, I do not believe that fiscal stimulus is ever really necessary. Thus my preference is for monetary stimulus, and I would favor moving to a system of price level or nominal GDP level targeting.
With respect to long run fiscal policy I favor a move to a progressive consumption tax as research evidence suggests that consumption taxes are growth enhancing and by being progressive it will address concerns about inequality. As far as spending goes I have no opposition to greater government spending as the research suggests that size of government has no statistically significant effect on long run growth one way or the other. In fact a more coherent system of social insurance such as they have in Europe would probably encourage the formation of human capital which according to the research evidence is itself growth enhancing.
P.S. Ever hear of Wagner’s Law?
Jeff
Mark,
Most of your lengthy response was beside the point. I was simply pointing out that the net effect of Richardo’s proposal of cutting taxes and cutting spending would contain within in it a 3% bump in GDP per Romer & Romer. And yet you seem to want to ignore this fact simply because the proposal would be classified as an endogenous observation. This argument just doesn’t hold any water.
2slugbaits,
The debt ceiling is not something new. It was first put in place in 1917, only four years after the federal income tax was imposed. Initially it was several ceilings, each applied to a different part of the budget. It was unified in 1939. What is new is not having it nearly automatically raised when it is approached.
I agree that if there is no agreement, it should be ignored. It is a stupid law, and no other country has ever had a nominal debt ceiling of this type. Time to retire the dumb thing.
Mark A. Sadowski
Jeff,
I think it’s very much to the point. A balanced-budget equivalent change in taxes and spending provides no fiscal stimulus at all. The Romers showed that once you exclude such changes the fiscal multiplier is much larger.
2slugbaits
Barkley Rosser My point was that for most of the country’s history we did not have a debt ceiling. There are still a few people alive today who were born before anyone ever thought of a debt ceiling. In other words, it’s not something that the Founders thought was necessary. In fact, the Reconstruction Founders wrote the 14th Amendment with the apparent belief that it served no good purpose. But we agree on the basic point…it’s a bad idea whose time has passed.
Rich Berger
I reread the Romers’ paper to clarify the difference between endogenous and exogenous tax changes. The former were those considered to return the economy to a normal growth path whereas the latter were those intended to cure a persistent deficit (e.g. Clinton 1993 tax increase) or spur long-term growth (Reagan 1981 and Bush 2003 tax cuts). The endogenous changes had minimal long-term effect, while the exogenous changes had the large effects noted above. Per the Romers tax increases had a large negative effect and tax cuts had a large positive effect on growth.
Ricardo
Mark wrote: I believe fiscal policy is effective in regulating aggregate demand.
So do I but that does not make it the right thing to do or that regulating aggregate demand will bring prosperity. Contrary to Keynesian beliefs aggregate demand tells us virtually nothing. For example the government was very successful in increasing aggregate demand in real estate prior to 2008. So in short I do not believe the discretionary fiscal stimulus has been ineffective. It’s just that we have a very serious recession which requires stronger action.
Your analysis simply assumes “a very serious recession” without looking at cause and effect. You do not understand Bastiat’s “Seen and Unseen” and so you praise the “seen” increase in demand without seeing the “unseen” capital destruction. In any case, as I do not believe in the liquidity trap, I do not believe that fiscal stimulus is ever really necessary.
An interesting break with Keynesian theory. Acutally the liquidity trap is real though not a Keynes postulated. The recent Great Recession has a clear manifestation of the liquidity trap. Look at excess bank reserves. Every time the FED attempts to “stimulate” with QE bank reserves grow. The transmission system through bank loans is broken because bad fiscal policy so monetary policy does virtually nothing but weaken the dollar. I favor a move to a progressive consumption tax … (UK 1970s) by being progressive it will address concerns about inequality.(US 1970s) I have no opposition to greater government spending(US 2008-current) a more coherent system of social insurance such as they have in Europe(Note unemployment rate in Europe since WWII and relate current US unemployment with a mature Social Security, Medicare, and Medicade system)
A very clear recipe for economic disaster. All of these things have been tried and ended in disaster. I have indicated representative failures after each of your points.
Wagner’s Law? Many economists throughout history have called it Leviathan. Adolph Wagner simply tried to put makeup on a pig.
Mark A. Sadowski
Rich Berger,
Yes I agree. But allow me to enlarge and clarify.
An endogenous tax change is designed to keep the economy at its normal growth path (balanced with spending changes) or to return it to the normal growth path (countercyclical and not balanced by spending changes). If it is designed to return it to the normal growth path it is a tax increase in the event the economy is overheated (e.g. the Tax Surcharge of 1968), or a tax decrease in the event the economy is in recession (e.g. the tax cut of 1975). If it is designed to keep it at the normal growth path it is balanced by a spending change (e.g. the Medicare tax increase of 1965 and the tax decrease of 1998). Implicit in this is the idea that any stimulative effect arises from a change in the budget balance.
Countercyclical tax changes probably fail to exhibit an effect because the effect is obscured by the condition of the economy. Tax changes balanced with spending changes fail to exhibit an effect because they do not change the budget balance.
That’s why a balanced-budget equivalent reduction in taxes and spending would have no effect.
Ricardo,
1) Other than the last few remaining adherents of Real Business Cycle Theory and a few eccentrics like Stephen Williamson there is no living professional macroeconomist who does not believe in the concept of aggregate demand. It is covered in virtually every principles class on macroeconomics on earth.
And aggregate demand is a macroeconomic concept. So it makes no sense to use it in the context of a single sector of the economy.
2) Some of what Bastiat said makes sense, but much does not. Interestingly, he was an advocate, with some reservations, of fiscal stimulus. Here is what Bastiat said about it in “What is Seen and What is Unseen”:
“There is an article in the Constitution which states:
‘Society assists and encourages the development of labor…. through the establishment by the state, the departments, and the municipalities, of appropriate public works to employ idle hands.’
As a temporary measure in a time of crisis, during a severe winter, this intervention on the part of the taxpayer could have good effects. It acts in the same way as insurance. It adds nothing to the number of jobs nor to total wages, but it takes labor and wages from ordinary times and doles them out, at a loss it is true, in difficult times.”
3) One reason reserves are so high right now is the interest on excess reserves currently being paid. It was instituted for the first time in Federal Reserve history on October 6, 2008. At 0.25% it currently pays more than 1,3 or 6 month T-bills and most commercial paper of terms of 90 days or less. Thus I have good reason to believe if we were not paying IOER that reserves would be much smaller.
Leaving that aside, in a low inflation environment, with large amounts of economic slack as we have right now, both the money multiplier and the velocity of money tend to decline, thus you may see a large increase in bank reserves. But that does not render monetary policy completely impotent.
Extreme examples of this are rare, such as the Great Depression in the US and the Lost Decade in Japan. During the Great Depression, after devaluing the dollar 41% in 1933, we had the swiftest growth in real GDP in any four year period in peacetime, an average of 9.5% annually during 1934-37. Similarly after Japan put ryōteki kin’yū kanwa (quantitative easing) into effect growth more than doubled from 0.9% on average during 1993-2002 to 2.4% on average during 2003-2007 and unemployment dropped for the first time in a decade.
Thus, in my opinion, the liquidity trap is a myth.
4) The List of Representative Failures
A) By my standards there has never been a progressive consumption tax, so with respect to the UK in the 1970s, I have no idea what you are talking about.
B) There was a sharp reduction in the top tax rate on earned income in 1971 and 1972 (from 70% to 60% to 50%) in the US so, again, I have no idea what you are talking about when you imply that taxes became more progressive in the 1970s. In any case progressivity in a consumption tax would exhibit very different economic effects than progressivity in an income tax. It would increase the incentives to save and invest.
C) Almost all of the recent increase in Federal spending in the US is related to the recession and so is temporary. In any case you are reversing cause and effect.
D) There is reason to believe that high levels of social insurance may be contributing to a higher rate of natural unemployment in some European countries. But that is hardly universal. For example Austria, Denmark, the Netherlands, Norway and Switzerland all traditionally have had lower unemployment rates than the US, and continue to do so today (currently 4.4%, 7.5%, 4.3%, 4.2% and 3.1% respectively). And all of these countries have very generous levels of social insurance.
How you can blame our current level of unemployment on Social Security, Medicare and Medicaid leaves me utterly perplexed.
In any case, I reiterate, the research evidence concerning the long run growth benefits of taxing consumption is quite strong. Similarly, on balance, the research evidence suggests that the level of government spending has no effect on long run growth.
Rich Berger
“Tax changes balanced with spending changes fail to exhibit an effect because they do not change the budget balance.
That’s why a balanced-budget equivalent reduction in taxes and spending would have no effect.”
Not buying it at all. As a reminder of the inefficacy of government spending see http://www.economics21.org/blog/revisiting-unemployment-predictions.
When you balance tax cuts with spending reductions, you are shifting resources from the unproductive to the productive sectors of the economy.
Rich Berger
The more I think about it, the more that I think a program of matched tax cuts and spending increases would be a winner for the country. According to the Romers, the GDP rises 2-3% for each 1% of GDP cut in taxes. Given that the Obama stimulus analysis assumed a multiplier of 1.55 (according to the e21 piece), there’s an arbtirage opportunity.
aaron
Rich, that’s great, for people with access to large amounts if cheap debt.
Rich Berger
Aaron-
You’re right. We have to get the growth up which will raise revenues under the current system of tax rates. And we have to change management ASAP.
Mark A. Sadowski
Rich Berger,
What you have there (e21) is evidence of why countercyclical tax cuts probably fail to exhibit an effect. The 2009 fiscal stimulus included $288 billion in tax cuts.
Ricardo
Mark wrote: Other than the last few remaining adherents of Real Business Cycle Theory and a few eccentrics like Stephen Williamson there is no living professional macroeconomist who does not believe in the concept of aggregate demand. It is covered in virtually every principles class on macroeconomics on earth.
Mark,
I will not address the bulk of your post because we have already been over it. But let me address the question of aggregate demand in academia.
Of course there is aggregate demand. If you take all of the demand by individuals and add it together you get aggregate demand, but the number is, as I said, virtually meaningless.
If you would like to meet some very distinguished economists who question the wisdom of Keynesian decision making using aggregate demand I will be glad to give you a whole host of names. You obviously need to get out more.
Only macroeconomists live in an aggregate world. The rest of the world recognizes that we have seasons of hot and cold and sometimes it makes sense to change clothes rather than dress for the aggregate.
Real economists recognize that principles of macroeconomics should be forgotten as soon as possible.
Rich Berger
Mark
Can you give any examples of a successful stimulus? Why did Romer and Bernstein say that the 2009 stimulus would work even with the (non aupply side) tax cuts?
2slugbaits
Rich Berger As R&R explain in their paper, the most obvious reason for the weak effects of countercyclical fiscal policies is the mistiming problem; i.e., in many cases the worst of the recession had largely passed by the time the tax changes went into effect. Surprisingly, R&R failed to recognize that even though the intent of those kinds of fiscal changes may have been “endogenous,” objective reality suggests that they were actually swimming in an “exogenous” sea despite the intent of the lawmakers. I think this is a weakness of the paper. You can also argue with some of their categorizations. For example, they classify the parts of the 2001 tax cut that were part of the 2000 campaign as being exogenous. And no doubt that’s how Team Bush would view it. But the actual rationale offered at the time was that the ballooning surplus would be a drag on the economy. So most Republicans as well as the Fed argued in 2000 that the proposed tax cut was really “endogenous.”
Finally, both you and Ricardo seem to have missed the main point of the R&R paper. It was a companion piece to an earlier paper they had done trying to find “exogenous” and “endogenous” changes in Fed policy. R&R were proposing a new methodology that they believed would give economists a new set of “instruments.” Since you probably aren’t familiar with this, let me give a simple example. We all know that supply and demand curves are drawn against quantity and price. What we actually observe are not supply and demand curves directly, but equilibrium quantities and prices. Connecting those equilibrium points (ordinarily) doesn’t give us a supply curve or a demand curve, just a set of equilibrium points. In order to tease out one of the curves you need some instrumental variable that is related to only one of the curves. That instrumental variable acts as a kind of anchor. One way to think of the R&R paper is that they are offering up narrative explanations of the motive for fiscal changes as a special kind of instrumental variable. That’s why they use “endogenous” and “exogenous” in somewhat unconventional ways.
Mark A. Sadowski
Ricardo,
I’ll save you some bother.
Barro, Kydland, Lucas, Plosser and Prescott are all distinguished macroeconomists that do not believe in the usefulness of the concept of aggregate demand. Sadly, Plosser is on the FOMC right now, which in my opinion, is like having the fox guard the henhouse.
There are others that have opposed both fiscal and monetary stimulus this recession but that is not because they do not believe in the usefulness of the concept of AD. If you take them at their word they actually believe the economy’s current performance is the best that can be expected.
Rich Berger,
1) I can and I have.
I gave two examples of what I believe are successful monetary stimulus above: 1) The devaluation of the dollar in 1933 and subsequent recovery in 1934-1937 and 2) quantitative easing in Japan in 2003-2007.
Fiscal stimulus is a little more problematic. You have to find a situation either where monetary policy is either focused on some other objective such as fixing the exchange rate, or where monetary authorities are for some reason passive in response (as they have been between QE1 and QE2). In the case of the US that would probably be the period between US entry into WW II and 1971-73 when we went to flexible exchange rates. I would argue that much of macroeconomic stabilization policy was done via tax and spending changes during that period.
In any case with flexible exchange rates the monetary authority is free to pursue aggregate demand management. (As I’ve said, I don’t believe in the liquidity trap.) Thus under such conditions I don’t believe fiscal stimulus is necessary or desirable. Someone else would probably make a better defense of fiscal stimulus.
2) Romer and Bernstein said it because they believed it. And I don’t really doubt it did what was forecast. (Furthermore demand side tax cuts were used because it was a fiscal stimulus, not a policy designed to encourage long run growth.)
In fact let’s refer back to the link you posted. Refer to the graph. At the time the original estimates were published (January 9th 2009) the unemployment rate was 6.8% (the December report did not come out until January 13th). The unemployment rate in 2008Q4 averaged 6.9%. According to Romer and Bernstein’s forecast they expected the unemployment rate to average 7.6% in 2009Q1, or an increase of 0.7 points. As it turned out unemployment actually average 8.2% or an increase of 1.3 points.
Now keep in mind that the first money from the fiscal stimulus didn’t come out until March and even then it was a trivial amount (checks to seniors). Thus the first effects of the stimulus were not really to be felt until 2009Q2.
So in short the unemployment rate was rising at nearly twice the rate that Romer and Bernstein had forecast and this occurred before the fiscal stimulus had a chance to really go into effect. It’s also reasonable to assume it would have taken longer for unemployment to stabilize in the absence of a fiscal stimulus. Since unemployment was originally forecast to go up to 9.0%, or by 2.1 points, we can suppose it probably would have actually gone up by much more.
Now this is just conjecture but follow me. Unemployment was rising at 1.86 times faster than forecast. Suppose that it took 1.86 times as long to stabilize. Then one should expect unemployment to go up by 1.86 squared fold more (3.45 fold).
So instead of 2.1 points we would be seeing an increase of 7.2 points or to 14.1%. Instead of the peak being reached in 2010Q1 (after 5 quarters) it would have been reached in 2011Q1 (after 9 quarters).
Now imagine this arc on the graph you posted. Then compare that result with what actually happened and note the vertical distance. Finally compare that distance with the distance that Romer and Bernstein estimated between the graph of with and without fiscal stimulus unemployment rates.
In short the recession was far more serious than forecast in early January 2009. Remember unemployment was only 6.8% then and would reach 8.6% before the fiscal stimulus was ever really implemented. The fiscal stimulus likely prevented something far, far worse.
In Romer’s defense she advocated a much larger plan, approximately $1.2 trillion. But it’s unlikely that a fiscal stimulus that size would have been passed.
Rich Berger
Mark-
I should have qualified that I meant a fiscal stimulus like the 2009 one. What amazes me is how little evidence there is that such a stimulus could be effective. If the evidence is so slight, how do they know that the multipliers are as large as they say they are? It’s also funny that the multipliers that they cite for tax cuts are so small, where the Romer paper indicates that the effects are much larger. I would call this faith-based economics.
Finally, I am skeptical of the aggregate demand paradigm too. As if demand were one undifferentiated mass and substituting one form of temporary demand for a real sustainable demand will do the trick. There is something fishy about the whole theory.
Ricardo
Mark,
I really appreciate you correcting the comment on aggregate demand. The economists you listed are a good start to a much longer list.
Mark wrote: There are others that have opposed both fiscal and monetary stimulus this recession but that is not because they do not believe in the usefulness of the concept of AD. If you take them at their word they actually believe the economy’s current performance is the best that can be expected.
You are correct here. Those who believe in aggregate demand theories yet oppose Keynesian fiscal and demand-side monetary stimulus are stuck. But there is significantly large and growing group of economists, most actually in business rather than academia, who are seeing that supply side, growth economics allows the economy to adjust and thrive while avoiding the bubbles and misallocations of capital that is endemic to a centrally planned, governmentally controlled system. You should check it out.
Ricardo
Slug,
I am very familiar with the Romers earlier paper on the effects of monetary policy, but the fact that their paper on tax changes is a companion to a paper on monetary changes does not change the facts reported in the papers.
Just for the record I believe that the Romers methodology and conclusions are highly suspect. The methodology requires a significant amount of subjective speculation and conjecture. Their papers are more opinion pieces than actual analysis.
The main reason I cite them is to try to speak in terms demand siders understand to try to help them arrive at supply side conclusions.
Rich Berger
There is an unreality about the whole idea of stimulus. In general, the problem is economic fluctuation, where plans by economic actors are shown to be flawed (markets for products and services shrinking) and cutbacks in investment, production and employment occur. Economic activity declines. The Keynesian approach is to create the illusion that demand has not fallen, rather than to let the adjustments run their course. The idea seems to be that you give the economy a kick start and once it’s running properly, you can stop kicking.
This has gotten far off Jim’s original post. I finally watched his interview yesterday and I think Jim provided some useful information to viewers without pushing a political agenda.
Mark A. Sadowski
Ricardo,
For the record I am not a demand sider. My dissertation topic is purely supply side in orientation (Tax Structure and Growth in the EU-27 during 1995-2007).
There’s no contradiction in being a supply sider AND appreciating demand side macroeconomics. (After all you wouldn’t treat a heart attack with chemotherapy, would you?)
P.S. Other than Barro, whose research is primarily on growth, the economists I named are all RBC. I consider RBC to be a cruel joke, whose total irrelevance has been revealed by the recent recession. This is very evident when you attend economics research seminars these days.
“The unemployed are merely displaying a preference for leisure.”
-Robert Lucas
Mark A. Sadowski
Ricardo,
One more thing. When you say “business rather then academia” be sure and check their crdentials. John Tamny calls himself a “Senior Economist” and preaches on macroeconomic issues (on taxes and monetary policy). But he’s never taken a single course in macroeconomics. In fact the only economics course he ever took was half a semester of “business economics” (which is essentially heavily watered down micro) when studying for his MBA in finance at Vanderbilt. Granted, he is reasonably well read (in 18th and 19th century economics literature).
2slugbaits
RicardoThe main reason I cite them is to try to speak in terms demand siders understand to try to help them arrive at supply side conclusions.
Except that I don’t think you are correctly understanding their findings. First, there are two examples of exogenous tax changes that they discuss in any detail. One was the Clinton 1993 tax increase to reduce the deficit. They claim that it did not have an adverse impact on output. The second exogenous tax change was the Kennedy/Johnson 1964 tax cut. This is the one that supply siders like to cite as an example of a growth oriented tax cut. And clearly it was, but the relevant question is whether it worked on the demand side or the supply side. Given that almost everyone agrees that a 90+% top marginal rate probably did discourage work effort, this tax cut has to be the gold standard for supply siders. Notice that R&R do not directly answer the question with respect to the 1964 tax cut. Instead they talk about the exogenous effects on output of a tax increase rather than a tax cut, with the implication that it works just the reverse for tax cuts. But the way they have to answer the question is by looking at tax increases. And what they found was that exogenous tax increases not dedicated to deficit reduction cut GDP growth. But here’s the important part and I direct you to the section that addresses Why Are The Effects So Persistent? R&R look at the behaviour of inflation and unemployment and conclude that there was no permanent effect on the supply side (i.e., productivity, labor hours, etc.), but there was a persistent deviation from normal output that is consistent with demand side effects. For example, the reduction in investment spending, which accounts for most of the GDP loss, is really just an example of the old Keynesian Flexible Accelerator model.
Bottom line: I don’t think you understood the paper. You misinterpreted it to claim that tax cuts contribute to supply side growth. That is not what the paper says. The paper says tax increases hurt growth…but for old fashioned Keynesian reasons, not supply side reasons.
Anonymous
“Why Are The Effects So Persistent? R&R look at the behaviour of inflation and unemployment and conclude that there was no permanent effect on the supply side (i.e., productivity, labor hours, etc.), but there was a persistent deviation from normal output that is consistent with demand side effects. For example, the reduction in investment spending, which accounts for most of the GDP loss, is really just an example of the old Keynesian Flexible Accelerator model.”
I reread that section and there is no support for your statement.
2slugbaits
Anonymous/RicardoI reread that section and there is no support for your statement.
Okay, I’ll finish your homework for you. This statement is professor-speak for talking about supply side effects: “Tax changes could have
large supply-side effects; that is, they could have large effects on the economy’s normal, flexible-price level of output.” This statement is professor-speak for aggregate demand responses: “Alternatively, tax changes could have highly persistent demand-side effects; that is, their impact on the gap between actual and normal output could be very long-lasting.”
I think that part is clear enough. Now, when the Romers talk about: “If the persistent output effects reflect changes in normal output, there is little reason to expect them to dissipate. If they reflect long-lasting departures of output from normal, on the other hand, one would expect them to fade.” This can be a little trickier to decipher. But the key phrase is “changes in normal output.” That phrase is talking about supply side effects; i.e., the Romers mean a shift in the supply curve away from the old normal. The words “long-lasting departures of output from normal” refers to demand side effects. In the paper the Romers sometimes use the word “departures” and sometimes “deviations,” but both words are in reference to demand side effects. With me so far? Good.
Now here’s the concluding paragraph for that section: “Thus, the behavior of inflation and unemployment are exactly what one would expect if the estimated long-lasting output effects of tax changes represented persistent departures of output from its flexible-price level. The evidence is not definitive, however: the inflation response is not precisely estimated, and it is conceivable that tax changes have large effects on the natural rate of unemployment.” In other words, they found the “departures from normal” explanation worked best. That is, they found stronger evidence for demand side effects. But note that the very last sentence does throw a bone to supply siders with the words “…it is conceivable that tax changes have large effects on the natural rate of unemployment.” The phrase “effects on the natural rate of unemployment” is, again, professor-speak for supply side effects.
I hope that cleared things up for you.
FWIW, I don’t think the R&R paper was particularly well written. I work for the government…worse yet, the Army, so I’m used to bad writing. I get plenty of practice every day; but the R&R paper is some pretty bad prose. I realize it was just a draft version, but I have a gated copy of the final version and it’s really not any better. One of my brothers is an English professor and he’d be banging his head against the wall if he had to read that kind of prose.
Ricardo
Mark,
If it is possible I would love to read your dissertation. Two of the most important works on the Great Depression I have ever read were dissertations. The candidates both took positions out of the mainstream but had more sound reasoning than anything I have seen written on the period. Let me know if it is possible.
Ricardo
Mark,
Let me give you an interesting name in academia and business, Professor Reuven Brenner.
Rich Berger
2sb-
That may have convinced you, but it didn’t convince me. A lot of hand waving.
2slugbaits
Rich Berger Okay, then let’s hear your version. Please explain what you think the Romers meant by “the economy’s normal, flexible-price level of output.” Please explain your understanding of “gap between actual and normal output.” Then tells us which curve is associated with “persistent departures of output from its flexible-price level.” Do you know what the Romers are talking about when they use the term “flexible-price level?” Finally, how do you interpret the term “the natural rate of unemployment.” Which curve is all about the “natural rate of unemployment?” I look forward to reading your reply.
Ricardo
Slug,
You do a great job of demonstrating that when the Romers talk about “supply side” they are actually talking about demand. I have found few demand side economists who understand supply side because theue entire methodology is wrapped up in consumption statistics.
This inability is usualy wrapped up in a lack of understanding of JB Say. He never said as Keynes and his followers loved to state that “supply creates its own demand.” This argument relied on reductive fallacy which was then turned on its head. If you do not understand the concept “one produces to consume” you are forever lost in a Bizarro World of effect preceeding cause, consumption preceeding production.
2slugbaits
Ricardo Nice try, but a red herring. We’re not talking about your understanding of Say’s Law. We weren’t discussing whether or not the Romers and Keynes correctly understood Say. The question on the table is whether or not you correctly understood the Romers’ paper.
Anonymous
Sure-
From the last sentence of the section you referenced:
“Thus, the behavior of inflation and unemployment are exactly what one would expect if the
estimated long-lasting output effects of tax changes represented persistent departures of output from its flexible-price level. The evidence is not definitive, however: the inflation response is not precisely estimated, and it is conceivable that tax changes have large effects on the natural rate of unemployment.”
Awful lot of hedging there, wouldn’t you agree? If they wanted to say that the effects of exogenous tax changes were not due to supply-side effects, why didn’t they just say so?
I think the idea that you can identify a level of output that is optimal or natural is suspect. In a sense, this is an assumption of Keynesian economics that serves a purpose similar to that of the ether pre-relativity. It cannot be observed, only estimated by models. And if the effects of exogenous tax changes are as persistent as they observe, isn’t the attribution to a non-supply side factor a distinction without a difference?
Even their attempt to characterize the effects of exo changes as deviations from the flexible-price output level is very convoluted, using Phillips curve methodology, and even so they say that “The one important caveat to this discussion is that the response of inflation to the tax cut is not precisely estimated. For example, there is only one quarter for which the null hypothesis of no effect can be rejected.”
They further say that a tax increase is followed by a large and significant increase in unemployment – certainly not what the supply side might expect, eh?
The more I read the more I think that the entire edifice of modern macroeconomic theory is rotten, with assumptions piled on assumptions, and complex mathematics employed to produce what? The economy, like the weather and climate, is too complex to allow detailed and precise prediction.
2slugbaits
Anonymous / Ricardo If you disagree with the paper, then fine. I have plenty of problems with it as well. But remember, you’re the one who brought it up. You’re the one who cited it as somehow supporting your viewpoint. I think we can wrap-up this whole long discussion with a simple summary: Ricardo thought that the Romer & Romer paper supported his viewpoint because one of the findings was that tax increases can sometimes hurt GDP growth and by extension tax cuts can sometimes help GDP growth. But upon further reflection Ricardo now believes that he was originally mistaken in his initial assessment of the R&R paper and that the authors were not saying what he thought they were saying. The poster Ricardo would therefore like to withdraw his previous statements and strike all references to the R&R paper from the record. Does that pretty well sum up where we’re at?
Rich Berger
No.
Ricardo
Thanks Rich, 😉 Slug is a hoot sometimes.
Rich Berger
Ricardo-
Agreed. I am sure there will be a future post where the fur flies again. I think this one has run its course.
The USA could benefit from overhauling the constitution and government. The current system reflects the highly stratified class society of the early industrial revolution period in Great Britain.
JDH The debt ceiling provision is (historically speaking) a fairly recent addition to the budget & execution process. For most of US history the Congressional authorization was all that was needed. If there comes a time when the Treasury is not able to meet creditors’ demands, my best guess is that the Treasury would ignore the debt ceiling and just continue to pay out. This would set up something of a constitutional “crisis,” but the debt ceiling provision is constitutionally dubious anyway.
I’m not sure I’m following your point about 2 Aug not being a hard date. If you’re saying that we don’t really know whether it might be 3 Aug or 1 Aug, then okay, I’ll grant you that. There’s always going to be some error around the estimate. But if you’re suggesting that the Treasury could really push things out weeks beyond 2 Aug, then I find that claim a little hard to swallow. Somebody is not going to get paid. [Interestingly, SS benefit checks are scheduled to go out on 3 Aug.] It would be a very bad negotiating tactic. If you threaten something, then you better follow through; and since executing the threat is something that the Administration does not want to do, the way you make that threat credible is by not giving yourself any alternative. So I think 2 Aug is the Treasury’s best estimate of a hard date, although they may have just enough wiggle room to make the 3 Aug SS benefit date.
But hard date or soft date, what really counts is what the markets think. If we’re in the 3rd week in July and the parties are no closer to agreement than they are now, then I don’t expect the whiz kids at Goldman-Sachs to just sit on their butts waiting to see if 2 Aug is really a hard date or not. The interesting complication is that late July may also be when the EU has to start dealing with some hard dates of their own regarding Europe’s periphery. What’s an investor to do?
Good Interview
Thanks,very precise and clear as always.The decor is medieval not matching the expected flamboyant gothic of our golden era.
Professor,
Great interview. It is hard to explain complex subjects in a couple of minutes and even then the talking head talks more than the person who actually has the expertise.
There are a minority of us who make the same point you did about a disconnect between spending, taxes, and debt. We recognize that the congress spends first and looks for a way to pay later whether taxes or debt. But the question is what are the parties doing to reconnect the three? What are you hearing from the Democrats? They want to ignore the impact of taxes and the out of control debt and not only keep spending high but even increase spending.
It seems that the Republicans actually are attempting to make the connection, though weakly, in their connecting spending cuts to increasing the debt ceiling. It is the Democrats who what to keep the two separate.
On taxes Christina and David Romer have given us the answer. A 1% increase in taxes means a 3% decline in GDP. Increasing taxes is a non-starter if you actually believe our problem is a lack of growth. They have also pointed out that a 1% decrease in taxes translates into a 2-3% increase in GDP.
If the deficit is going to increase due to spending or to tax cuts doesn’t it make a lot more sense to decrease taxes and get a 3% increase in GDP rather than maintaining high spending and raising taxes causing a 2-3% decrease in our anemic 1.8% growth. Doesn’t that mean that taxing will actually push us into GDP decline?
The plan is actually very simple. Cut taxes to gain that 3% growth, cut spending to return to sanity concerning the deficit, lock the debt limit to no additional increases so that 1 year from now we are not back dealing with this issue like we did in January 2010.
Ricardo,
It is my understanding, based on things she and David have said, that Christina Romer believes that the tax multiplier is no larger than the spending multiplier. And she certainly does not believe that a balanced-budget equivalent reduction in taxes and spending provides any fiscal stimulus at all.
To compare the fiscal multipliers estimated previously with the Romers’ estimate would be inappropriate. They used a very different (and in some ways revolutionary) methodology. Thus it is not comparable. For example, previous studies have not fully controlled for the tax increases that often accompany spending increases. Thus they are very likely to understate the effects of spending increases alone. Valerie Ramey’s study in particular assessed the impact of the Korean-War military spending increase without taking account of the fact that it was accompanied by a large tax increase.
What the Romers’ study showed is not that tax cuts are uniquely effective, but rather that failing to consider the reasons for tax policy changes leads to an underestimation of their effects. Because these issues of omitted variable bias are likely to be as strong for spending as for tax changes, the most reasonable interpretation of her paper is that all types of fiscal stimulus are more potent than conventional estimates would lead us to believe.
A passage of the paper that helps to drive the correct interpretation of the paper home is the following:
“Following a spending-driven tax increase, real GDP on average rises moderately, reaching a maximum of 0.7 percent after two quarters (t = 1.5). Thereafter the effect fluctuates irregularly around zero and is always far from significant. Thus looking at how the economy behaves after tax changes driven by spending changes yields estimates of the effects of tax changes that are starkly different from those based on exogenous tax changes.”
In short tax increases coupled with spending increases yielded no statistically significant effect on the economy, and in fact, in direct contradiction to unrealistic supply side enthusiasts’ beliefs, on average, at least temporarily, they increased GDP. (I have no problem with the idea that taxes have a long run effect on AS, but we are of course talking about the short run effect on AD here.)
In the final analysis, the Romers’ paper found that, with the exception of countercyclical tax changes (possibly because there were too few, or because the economic downturn obscured their effect), the only type of tax change that had a statistically significant effect on GDP were those that changed the federal budget balance. This is hardly a result that supports your plan of action.
Mark A. Sadowski Thanks for doing the heavy lifting. It saved me a lot of effort. Ricardo has brought up the Romer & Romer paper several times, and Menzie has politely suggested to Ricardo that he might want to reread their paper (assuming he has actually read it in the first place). As you pointed out, the Romer & Romer paper actually undercuts Ricardo’s argument. And like a lot of folks who post here, Ricardo doesn’t understand the difference between an AD curve and an AS curve. Today all of the action (or inaction) is on the AD curve. Ricardo still thinks it’s 1979.
ISTM that in the face of Congressional guidance to spend X, Tax Y, and borrow no more than Z, the only legal option left to the administration is to sell assets. There’s all that yellow stuff in Kentucky, that might allow the US to operate a few months more…
Romer paper is at http://elsa.berkeley.edu/~cromer/RomerDraft307.pdf
In Section VI Conclusions is the following excerpt:
“In terms of consequences, there are six main findings. First, tax changes have very large effects on output. Our baseline specification suggests that an exogenous tax increase of one percent of GDP lowers real GDP by roughly three percent. Our many robustness checks for the most part point to a slightly smaller decline, but one that is still well over two percent. Second, these estimated effects are substantially larger than those obtained using broader measures of tax changes, such as the change in cyclically adjusted revenues or all legislated tax changes. This suggests that failing to account for the
reasons for tax changes can lead to substantially biased estimates of the macroeconomic effects of fiscal actions. Third, investment falls sharply in response to exogenous tax increases. Indeed, the strong response of investment helps to explain why the output consequences of tax changes are so large. Fourth, the output effects of tax changes are highly persistent. The behavior of inflation and unemployment
suggests that this persistence reflects long-lasting departures of output from its flexible-price level, not large effects of tax changes on the flexible-price level of output.”
Sounds like Ricardo quoted the paper very accurately.
Mark and Slug,
What the Great Depression, the Great Inflation, and the current Great Recession have proved and are proving for the umpteenth time is that the so-called spending multipliers don’t work at best and probably create bubbles to make things worse. Now that we have followed Christina and David’s spending multiplier idea to abject failure why not try the tax cut multiplier and see if it works again:
as it did in the UK after the Napoleonic Wars,
as it did in the US and Italy after WWI,
as it did in Germany and the US after WWII,
as it did with the Kennedy-Johnson tax cuts in the 1960s,
as it did with the Reagan tax cuts,
as it did with Clinton’s capital gains tax cuts,
Just to mention a few.
Anonymous,
The issue is not the estimated effect of an exogenous tax change. The issue is whether the Romers’ research implies that tax multipliers have a different effect from spending multipliers. I think it’s quite clear from their results concerning endogenous tax changes that this is not the case. In the final analysis the tax change Ricardo is proposing would not even be classified as exogenous by the Romers.
Ricardo,
Five of the eight historical incidents you list were examined by the Romers’ paper. And in most cases your anecdotal evidence involves simultaneous spending changes and is conditional on the monetary policy response function.
From the Abstract to the Romers’ article:
“This paper investigates the impact of changes in the level of taxation on economic activity. The paper uses the narrative record presidential speeches, executive-branch documents, and Congressional reports to identify the size, timing, and principal motivation for all major postwar tax policy actions. This narrative analysis allows us to separate revenue changes resulting from legislation from changes occurring for other reasons. It also allows us to further separate legislated changes into those taken for
reasons related to prospective economic conditions, such as countercyclical actions and tax changes tied to changes in government spending, and those taken for more exogenous reasons, such as to reduce an inherited budget deficit or to promote long-run growth. We then examine the behavior of output following these more exogenous legislated changes. The resulting estimates indicate that tax increases are highly contractionary. The effects are strongly significant, highly robust, and much larger than those obtained using broader measures of tax changes. The large effect stems in considerable part from a powerful negative effect of tax increases on investment. We also find that legislated tax increases designed to reduce a persistent budget deficit appear to have much smaller output costs than other tax
increases.”
They distinguish between endogenous and exogenous tax increases. As I understand the distinction, endogenous tax increases are those that follow through the natural working of the tax system (e.g., increasing incomes generate higher taxes) versus exogenous increases which are the result of deliberate legislation. They seem to say very unambiguously that the exogenous tax increases (such as those embodied in CBO’s extended baseline and alternative scenarios) are contractionary. The best that can be said is that they are less contractionary when devoted to reduction of persistent deficits. I think their paper provides additional ammunition in favor of the Ryan plan which does not require additional tax increases.
Incidentally, in a search of this site, I found a post back in 2007 where Prof Hamilton commented on this study and which was very helpful to me in understanding its basic thrust.
In the final analysis the tax change Ricardo is proposing would not even be classified as exogenous by the Romers.
Mark,
Once the effect of an explanatory variable has been determined, that effect does not change simply because a new observation is determined to be endogenous.
Mark wrote:
Ricardo,
Five of the eight historical incidents you list were examined by the Romers’ paper. And in most cases your anecdotal evidence involves simultaneous spending changes and is conditional on the monetary policy response function.
Mark,
True, but the Romers do not say that tax decreases were not part of the growth. What they said is that there are other contributing factors. So, once again I ask, why not try it? At the worst it seems that the Romers are saying it may have little to no effect, but they are not denying that it could have a great effect.
When we know one thing has failed but another has worked, why resist the one that works?
Rich Berger,
You wrote:
“As I understand the distinction, endogenous tax increases are those that follow through the natural working of the tax system (e.g., increasing incomes generate higher taxes) versus exogenous increases which are the result of deliberate legislation.”
Your understanding is incorrect. Both endogenous and exogenous tax changes are in fact the result of deliberate legislation. If you read the paper you’ll see they spent a great deal of time going through the legislative record classifying tax changes as one or the other. They state quite clearly that they specifically wanted to avoid changes in revenue that were purely the result of changes in income.
Endogenous tax changes include both countercyclical tax changes and tax changes coupled to spending changes (so as to minimize the change in budget balance). Exogenous tax changes are essentially all other tax changes.
Jeff,
You wrote:
“Once the effect of an explanatory variable has been determined, that effect does not change simply because a new observation is determined to be endogenous.”
The Romer’s motive was to isolate tax changes that changed a budget balance but which were not in response to economic conditions. Their concern was that the effect of tax changes taken in response to economic conditions might have their effect obscured by those same conditions.
Thus the econometric endogeneity in this context refers to whether or not it was in response to economic conditions. The fact that tax changes coupled with spending changes were classified as “endogenous” was merely to separate them from noncountercyclical tax changes that changed a budget balance.
The key to understanding this paper is too keep in mind it was done in the context of research on macroeconomic stabilization policy (short run growth). They were interested in detecting more clearly the effect of a change in the budget balance on aggregate expenditures by isolating it from economic conditions that might have obscured that effect. The rationale for working with legislated tax changes rather than spending changes seems to be that the data was easier to work with, although in theory the same methods could be applied to spending changes. In short it is part of the larger body of New Keynesian literature.
There is a small body of literature on the effects of tax level and tax structure on long run growth but this is not part of it. (I’m intimately familiar with it, as it forms the literature review to my dissertation.) That is part of the larger body of Growth literature.
Those trying to read supply side results into a paper that is demand side in focus are quite simply barking up the wrong tree.
Ricardo,
To the extent that monetary policy accomodates fiscal policy, I believe fiscal policy is effective in regulating aggregate demand. Between QE1 and QE2 the Fed did very little one way or another so I have no reason to doubt that the effects of the discretionary fiscal stimulus have been what the major private and public forecasters have estimated.
So in short I do not believe the discretionary fiscal stimulus has been ineffective. It’s just that we have a very serious recession which requires stronger action. And I certainly do not believe that the tax multiplier is any larger than the spending multiplier.
In any case, as I do not believe in the liquidity trap, I do not believe that fiscal stimulus is ever really necessary. Thus my preference is for monetary stimulus, and I would favor moving to a system of price level or nominal GDP level targeting.
With respect to long run fiscal policy I favor a move to a progressive consumption tax as research evidence suggests that consumption taxes are growth enhancing and by being progressive it will address concerns about inequality. As far as spending goes I have no opposition to greater government spending as the research suggests that size of government has no statistically significant effect on long run growth one way or the other. In fact a more coherent system of social insurance such as they have in Europe would probably encourage the formation of human capital which according to the research evidence is itself growth enhancing.
P.S. Ever hear of Wagner’s Law?
Mark,
Most of your lengthy response was beside the point. I was simply pointing out that the net effect of Richardo’s proposal of cutting taxes and cutting spending would contain within in it a 3% bump in GDP per Romer & Romer. And yet you seem to want to ignore this fact simply because the proposal would be classified as an endogenous observation. This argument just doesn’t hold any water.
2slugbaits,
The debt ceiling is not something new. It was first put in place in 1917, only four years after the federal income tax was imposed. Initially it was several ceilings, each applied to a different part of the budget. It was unified in 1939. What is new is not having it nearly automatically raised when it is approached.
I agree that if there is no agreement, it should be ignored. It is a stupid law, and no other country has ever had a nominal debt ceiling of this type. Time to retire the dumb thing.
Jeff,
I think it’s very much to the point. A balanced-budget equivalent change in taxes and spending provides no fiscal stimulus at all. The Romers showed that once you exclude such changes the fiscal multiplier is much larger.
Barkley Rosser My point was that for most of the country’s history we did not have a debt ceiling. There are still a few people alive today who were born before anyone ever thought of a debt ceiling. In other words, it’s not something that the Founders thought was necessary. In fact, the Reconstruction Founders wrote the 14th Amendment with the apparent belief that it served no good purpose. But we agree on the basic point…it’s a bad idea whose time has passed.
I reread the Romers’ paper to clarify the difference between endogenous and exogenous tax changes. The former were those considered to return the economy to a normal growth path whereas the latter were those intended to cure a persistent deficit (e.g. Clinton 1993 tax increase) or spur long-term growth (Reagan 1981 and Bush 2003 tax cuts). The endogenous changes had minimal long-term effect, while the exogenous changes had the large effects noted above. Per the Romers tax increases had a large negative effect and tax cuts had a large positive effect on growth.
Mark wrote:
I believe fiscal policy is effective in regulating aggregate demand.
So do I but that does not make it the right thing to do or that regulating aggregate demand will bring prosperity. Contrary to Keynesian beliefs aggregate demand tells us virtually nothing. For example the government was very successful in increasing aggregate demand in real estate prior to 2008.
So in short I do not believe the discretionary fiscal stimulus has been ineffective. It’s just that we have a very serious recession which requires stronger action.
Your analysis simply assumes “a very serious recession” without looking at cause and effect. You do not understand Bastiat’s “Seen and Unseen” and so you praise the “seen” increase in demand without seeing the “unseen” capital destruction.
In any case, as I do not believe in the liquidity trap, I do not believe that fiscal stimulus is ever really necessary.
An interesting break with Keynesian theory. Acutally the liquidity trap is real though not a Keynes postulated. The recent Great Recession has a clear manifestation of the liquidity trap. Look at excess bank reserves. Every time the FED attempts to “stimulate” with QE bank reserves grow. The transmission system through bank loans is broken because bad fiscal policy so monetary policy does virtually nothing but weaken the dollar.
I favor a move to a progressive consumption tax … (UK 1970s)
by being progressive it will address concerns about inequality.(US 1970s)
I have no opposition to greater government spending(US 2008-current)
a more coherent system of social insurance such as they have in Europe(Note unemployment rate in Europe since WWII and relate current US unemployment with a mature Social Security, Medicare, and Medicade system)
A very clear recipe for economic disaster. All of these things have been tried and ended in disaster. I have indicated representative failures after each of your points.
Wagner’s Law? Many economists throughout history have called it Leviathan. Adolph Wagner simply tried to put makeup on a pig.
Rich Berger,
Yes I agree. But allow me to enlarge and clarify.
An endogenous tax change is designed to keep the economy at its normal growth path (balanced with spending changes) or to return it to the normal growth path (countercyclical and not balanced by spending changes). If it is designed to return it to the normal growth path it is a tax increase in the event the economy is overheated (e.g. the Tax Surcharge of 1968), or a tax decrease in the event the economy is in recession (e.g. the tax cut of 1975). If it is designed to keep it at the normal growth path it is balanced by a spending change (e.g. the Medicare tax increase of 1965 and the tax decrease of 1998). Implicit in this is the idea that any stimulative effect arises from a change in the budget balance.
Countercyclical tax changes probably fail to exhibit an effect because the effect is obscured by the condition of the economy. Tax changes balanced with spending changes fail to exhibit an effect because they do not change the budget balance.
That’s why a balanced-budget equivalent reduction in taxes and spending would have no effect.
Ricardo,
1) Other than the last few remaining adherents of Real Business Cycle Theory and a few eccentrics like Stephen Williamson there is no living professional macroeconomist who does not believe in the concept of aggregate demand. It is covered in virtually every principles class on macroeconomics on earth.
And aggregate demand is a macroeconomic concept. So it makes no sense to use it in the context of a single sector of the economy.
2) Some of what Bastiat said makes sense, but much does not. Interestingly, he was an advocate, with some reservations, of fiscal stimulus. Here is what Bastiat said about it in “What is Seen and What is Unseen”:
“There is an article in the Constitution which states:
‘Society assists and encourages the development of labor…. through the establishment by the state, the departments, and the municipalities, of appropriate public works to employ idle hands.’
As a temporary measure in a time of crisis, during a severe winter, this intervention on the part of the taxpayer could have good effects. It acts in the same way as insurance. It adds nothing to the number of jobs nor to total wages, but it takes labor and wages from ordinary times and doles them out, at a loss it is true, in difficult times.”
3) One reason reserves are so high right now is the interest on excess reserves currently being paid. It was instituted for the first time in Federal Reserve history on October 6, 2008. At 0.25% it currently pays more than 1,3 or 6 month T-bills and most commercial paper of terms of 90 days or less. Thus I have good reason to believe if we were not paying IOER that reserves would be much smaller.
Leaving that aside, in a low inflation environment, with large amounts of economic slack as we have right now, both the money multiplier and the velocity of money tend to decline, thus you may see a large increase in bank reserves. But that does not render monetary policy completely impotent.
Extreme examples of this are rare, such as the Great Depression in the US and the Lost Decade in Japan. During the Great Depression, after devaluing the dollar 41% in 1933, we had the swiftest growth in real GDP in any four year period in peacetime, an average of 9.5% annually during 1934-37. Similarly after Japan put ryōteki kin’yū kanwa (quantitative easing) into effect growth more than doubled from 0.9% on average during 1993-2002 to 2.4% on average during 2003-2007 and unemployment dropped for the first time in a decade.
Thus, in my opinion, the liquidity trap is a myth.
4) The List of Representative Failures
A) By my standards there has never been a progressive consumption tax, so with respect to the UK in the 1970s, I have no idea what you are talking about.
B) There was a sharp reduction in the top tax rate on earned income in 1971 and 1972 (from 70% to 60% to 50%) in the US so, again, I have no idea what you are talking about when you imply that taxes became more progressive in the 1970s. In any case progressivity in a consumption tax would exhibit very different economic effects than progressivity in an income tax. It would increase the incentives to save and invest.
C) Almost all of the recent increase in Federal spending in the US is related to the recession and so is temporary. In any case you are reversing cause and effect.
D) There is reason to believe that high levels of social insurance may be contributing to a higher rate of natural unemployment in some European countries. But that is hardly universal. For example Austria, Denmark, the Netherlands, Norway and Switzerland all traditionally have had lower unemployment rates than the US, and continue to do so today (currently 4.4%, 7.5%, 4.3%, 4.2% and 3.1% respectively). And all of these countries have very generous levels of social insurance.
How you can blame our current level of unemployment on Social Security, Medicare and Medicaid leaves me utterly perplexed.
In any case, I reiterate, the research evidence concerning the long run growth benefits of taxing consumption is quite strong. Similarly, on balance, the research evidence suggests that the level of government spending has no effect on long run growth.
“Tax changes balanced with spending changes fail to exhibit an effect because they do not change the budget balance.
That’s why a balanced-budget equivalent reduction in taxes and spending would have no effect.”
Not buying it at all. As a reminder of the inefficacy of government spending see http://www.economics21.org/blog/revisiting-unemployment-predictions.
When you balance tax cuts with spending reductions, you are shifting resources from the unproductive to the productive sectors of the economy.
The more I think about it, the more that I think a program of matched tax cuts and spending increases would be a winner for the country. According to the Romers, the GDP rises 2-3% for each 1% of GDP cut in taxes. Given that the Obama stimulus analysis assumed a multiplier of 1.55 (according to the e21 piece), there’s an arbtirage opportunity.
Rich, that’s great, for people with access to large amounts if cheap debt.
Aaron-
You’re right. We have to get the growth up which will raise revenues under the current system of tax rates. And we have to change management ASAP.
Rich Berger,
What you have there (e21) is evidence of why countercyclical tax cuts probably fail to exhibit an effect. The 2009 fiscal stimulus included $288 billion in tax cuts.
Mark wrote:
Other than the last few remaining adherents of Real Business Cycle Theory and a few eccentrics like Stephen Williamson there is no living professional macroeconomist who does not believe in the concept of aggregate demand. It is covered in virtually every principles class on macroeconomics on earth.
Mark,
I will not address the bulk of your post because we have already been over it. But let me address the question of aggregate demand in academia.
Of course there is aggregate demand. If you take all of the demand by individuals and add it together you get aggregate demand, but the number is, as I said, virtually meaningless.
If you would like to meet some very distinguished economists who question the wisdom of Keynesian decision making using aggregate demand I will be glad to give you a whole host of names. You obviously need to get out more.
Only macroeconomists live in an aggregate world. The rest of the world recognizes that we have seasons of hot and cold and sometimes it makes sense to change clothes rather than dress for the aggregate.
Real economists recognize that principles of macroeconomics should be forgotten as soon as possible.
Mark
Can you give any examples of a successful stimulus? Why did Romer and Bernstein say that the 2009 stimulus would work even with the (non aupply side) tax cuts?
Rich Berger As R&R explain in their paper, the most obvious reason for the weak effects of countercyclical fiscal policies is the mistiming problem; i.e., in many cases the worst of the recession had largely passed by the time the tax changes went into effect. Surprisingly, R&R failed to recognize that even though the intent of those kinds of fiscal changes may have been “endogenous,” objective reality suggests that they were actually swimming in an “exogenous” sea despite the intent of the lawmakers. I think this is a weakness of the paper. You can also argue with some of their categorizations. For example, they classify the parts of the 2001 tax cut that were part of the 2000 campaign as being exogenous. And no doubt that’s how Team Bush would view it. But the actual rationale offered at the time was that the ballooning surplus would be a drag on the economy. So most Republicans as well as the Fed argued in 2000 that the proposed tax cut was really “endogenous.”
Finally, both you and Ricardo seem to have missed the main point of the R&R paper. It was a companion piece to an earlier paper they had done trying to find “exogenous” and “endogenous” changes in Fed policy. R&R were proposing a new methodology that they believed would give economists a new set of “instruments.” Since you probably aren’t familiar with this, let me give a simple example. We all know that supply and demand curves are drawn against quantity and price. What we actually observe are not supply and demand curves directly, but equilibrium quantities and prices. Connecting those equilibrium points (ordinarily) doesn’t give us a supply curve or a demand curve, just a set of equilibrium points. In order to tease out one of the curves you need some instrumental variable that is related to only one of the curves. That instrumental variable acts as a kind of anchor. One way to think of the R&R paper is that they are offering up narrative explanations of the motive for fiscal changes as a special kind of instrumental variable. That’s why they use “endogenous” and “exogenous” in somewhat unconventional ways.
Ricardo,
I’ll save you some bother.
Barro, Kydland, Lucas, Plosser and Prescott are all distinguished macroeconomists that do not believe in the usefulness of the concept of aggregate demand. Sadly, Plosser is on the FOMC right now, which in my opinion, is like having the fox guard the henhouse.
There are others that have opposed both fiscal and monetary stimulus this recession but that is not because they do not believe in the usefulness of the concept of AD. If you take them at their word they actually believe the economy’s current performance is the best that can be expected.
Rich Berger,
1) I can and I have.
I gave two examples of what I believe are successful monetary stimulus above: 1) The devaluation of the dollar in 1933 and subsequent recovery in 1934-1937 and 2) quantitative easing in Japan in 2003-2007.
Fiscal stimulus is a little more problematic. You have to find a situation either where monetary policy is either focused on some other objective such as fixing the exchange rate, or where monetary authorities are for some reason passive in response (as they have been between QE1 and QE2). In the case of the US that would probably be the period between US entry into WW II and 1971-73 when we went to flexible exchange rates. I would argue that much of macroeconomic stabilization policy was done via tax and spending changes during that period.
In any case with flexible exchange rates the monetary authority is free to pursue aggregate demand management. (As I’ve said, I don’t believe in the liquidity trap.) Thus under such conditions I don’t believe fiscal stimulus is necessary or desirable. Someone else would probably make a better defense of fiscal stimulus.
2) Romer and Bernstein said it because they believed it. And I don’t really doubt it did what was forecast. (Furthermore demand side tax cuts were used because it was a fiscal stimulus, not a policy designed to encourage long run growth.)
In fact let’s refer back to the link you posted. Refer to the graph. At the time the original estimates were published (January 9th 2009) the unemployment rate was 6.8% (the December report did not come out until January 13th). The unemployment rate in 2008Q4 averaged 6.9%. According to Romer and Bernstein’s forecast they expected the unemployment rate to average 7.6% in 2009Q1, or an increase of 0.7 points. As it turned out unemployment actually average 8.2% or an increase of 1.3 points.
Now keep in mind that the first money from the fiscal stimulus didn’t come out until March and even then it was a trivial amount (checks to seniors). Thus the first effects of the stimulus were not really to be felt until 2009Q2.
So in short the unemployment rate was rising at nearly twice the rate that Romer and Bernstein had forecast and this occurred before the fiscal stimulus had a chance to really go into effect. It’s also reasonable to assume it would have taken longer for unemployment to stabilize in the absence of a fiscal stimulus. Since unemployment was originally forecast to go up to 9.0%, or by 2.1 points, we can suppose it probably would have actually gone up by much more.
Now this is just conjecture but follow me. Unemployment was rising at 1.86 times faster than forecast. Suppose that it took 1.86 times as long to stabilize. Then one should expect unemployment to go up by 1.86 squared fold more (3.45 fold).
So instead of 2.1 points we would be seeing an increase of 7.2 points or to 14.1%. Instead of the peak being reached in 2010Q1 (after 5 quarters) it would have been reached in 2011Q1 (after 9 quarters).
Now imagine this arc on the graph you posted. Then compare that result with what actually happened and note the vertical distance. Finally compare that distance with the distance that Romer and Bernstein estimated between the graph of with and without fiscal stimulus unemployment rates.
In short the recession was far more serious than forecast in early January 2009. Remember unemployment was only 6.8% then and would reach 8.6% before the fiscal stimulus was ever really implemented. The fiscal stimulus likely prevented something far, far worse.
In Romer’s defense she advocated a much larger plan, approximately $1.2 trillion. But it’s unlikely that a fiscal stimulus that size would have been passed.
Mark-
I should have qualified that I meant a fiscal stimulus like the 2009 one. What amazes me is how little evidence there is that such a stimulus could be effective. If the evidence is so slight, how do they know that the multipliers are as large as they say they are? It’s also funny that the multipliers that they cite for tax cuts are so small, where the Romer paper indicates that the effects are much larger. I would call this faith-based economics.
Finally, I am skeptical of the aggregate demand paradigm too. As if demand were one undifferentiated mass and substituting one form of temporary demand for a real sustainable demand will do the trick. There is something fishy about the whole theory.
Mark,
I really appreciate you correcting the comment on aggregate demand. The economists you listed are a good start to a much longer list.
Mark wrote:
There are others that have opposed both fiscal and monetary stimulus this recession but that is not because they do not believe in the usefulness of the concept of AD. If you take them at their word they actually believe the economy’s current performance is the best that can be expected.
You are correct here. Those who believe in aggregate demand theories yet oppose Keynesian fiscal and demand-side monetary stimulus are stuck. But there is significantly large and growing group of economists, most actually in business rather than academia, who are seeing that supply side, growth economics allows the economy to adjust and thrive while avoiding the bubbles and misallocations of capital that is endemic to a centrally planned, governmentally controlled system. You should check it out.
Slug,
I am very familiar with the Romers earlier paper on the effects of monetary policy, but the fact that their paper on tax changes is a companion to a paper on monetary changes does not change the facts reported in the papers.
Just for the record I believe that the Romers methodology and conclusions are highly suspect. The methodology requires a significant amount of subjective speculation and conjecture. Their papers are more opinion pieces than actual analysis.
The main reason I cite them is to try to speak in terms demand siders understand to try to help them arrive at supply side conclusions.
There is an unreality about the whole idea of stimulus. In general, the problem is economic fluctuation, where plans by economic actors are shown to be flawed (markets for products and services shrinking) and cutbacks in investment, production and employment occur. Economic activity declines. The Keynesian approach is to create the illusion that demand has not fallen, rather than to let the adjustments run their course. The idea seems to be that you give the economy a kick start and once it’s running properly, you can stop kicking.
This has gotten far off Jim’s original post. I finally watched his interview yesterday and I think Jim provided some useful information to viewers without pushing a political agenda.
Ricardo,
For the record I am not a demand sider. My dissertation topic is purely supply side in orientation (Tax Structure and Growth in the EU-27 during 1995-2007).
There’s no contradiction in being a supply sider AND appreciating demand side macroeconomics. (After all you wouldn’t treat a heart attack with chemotherapy, would you?)
P.S. Other than Barro, whose research is primarily on growth, the economists I named are all RBC. I consider RBC to be a cruel joke, whose total irrelevance has been revealed by the recent recession. This is very evident when you attend economics research seminars these days.
“The unemployed are merely displaying a preference for leisure.”
-Robert Lucas
Ricardo,
One more thing. When you say “business rather then academia” be sure and check their crdentials. John Tamny calls himself a “Senior Economist” and preaches on macroeconomic issues (on taxes and monetary policy). But he’s never taken a single course in macroeconomics. In fact the only economics course he ever took was half a semester of “business economics” (which is essentially heavily watered down micro) when studying for his MBA in finance at Vanderbilt. Granted, he is reasonably well read (in 18th and 19th century economics literature).
Ricardo The main reason I cite them is to try to speak in terms demand siders understand to try to help them arrive at supply side conclusions.
Except that I don’t think you are correctly understanding their findings. First, there are two examples of exogenous tax changes that they discuss in any detail. One was the Clinton 1993 tax increase to reduce the deficit. They claim that it did not have an adverse impact on output. The second exogenous tax change was the Kennedy/Johnson 1964 tax cut. This is the one that supply siders like to cite as an example of a growth oriented tax cut. And clearly it was, but the relevant question is whether it worked on the demand side or the supply side. Given that almost everyone agrees that a 90+% top marginal rate probably did discourage work effort, this tax cut has to be the gold standard for supply siders. Notice that R&R do not directly answer the question with respect to the 1964 tax cut. Instead they talk about the exogenous effects on output of a tax increase rather than a tax cut, with the implication that it works just the reverse for tax cuts. But the way they have to answer the question is by looking at tax increases. And what they found was that exogenous tax increases not dedicated to deficit reduction cut GDP growth. But here’s the important part and I direct you to the section that addresses Why Are The Effects So Persistent? R&R look at the behaviour of inflation and unemployment and conclude that there was no permanent effect on the supply side (i.e., productivity, labor hours, etc.), but there was a persistent deviation from normal output that is consistent with demand side effects. For example, the reduction in investment spending, which accounts for most of the GDP loss, is really just an example of the old Keynesian Flexible Accelerator model.
Bottom line: I don’t think you understood the paper. You misinterpreted it to claim that tax cuts contribute to supply side growth. That is not what the paper says. The paper says tax increases hurt growth…but for old fashioned Keynesian reasons, not supply side reasons.
“Why Are The Effects So Persistent? R&R look at the behaviour of inflation and unemployment and conclude that there was no permanent effect on the supply side (i.e., productivity, labor hours, etc.), but there was a persistent deviation from normal output that is consistent with demand side effects. For example, the reduction in investment spending, which accounts for most of the GDP loss, is really just an example of the old Keynesian Flexible Accelerator model.”
I reread that section and there is no support for your statement.
Anonymous/Ricardo I reread that section and there is no support for your statement.
Okay, I’ll finish your homework for you. This statement is professor-speak for talking about supply side effects: “Tax changes could have
large supply-side effects; that is, they could have large effects on the economy’s normal, flexible-price level of output.” This statement is professor-speak for aggregate demand responses: “Alternatively, tax changes could have highly persistent demand-side effects; that is, their impact on the gap between actual and normal output could be very long-lasting.”
I think that part is clear enough. Now, when the Romers talk about: “If the persistent output effects reflect changes in normal output, there is little reason to expect them to dissipate. If they reflect long-lasting departures of output from normal, on the other hand, one would expect them to fade.” This can be a little trickier to decipher. But the key phrase is “changes in normal output.” That phrase is talking about supply side effects; i.e., the Romers mean a shift in the supply curve away from the old normal. The words “long-lasting departures of output from normal” refers to demand side effects. In the paper the Romers sometimes use the word “departures” and sometimes “deviations,” but both words are in reference to demand side effects. With me so far? Good.
Now here’s the concluding paragraph for that section: “Thus, the behavior of inflation and unemployment are exactly what one would expect if the estimated long-lasting output effects of tax changes represented persistent departures of output from its flexible-price level. The evidence is not definitive, however: the inflation response is not precisely estimated, and it is conceivable that tax changes have large effects on the natural rate of unemployment.” In other words, they found the “departures from normal” explanation worked best. That is, they found stronger evidence for demand side effects. But note that the very last sentence does throw a bone to supply siders with the words “…it is conceivable that tax changes have large effects on the natural rate of unemployment.” The phrase “effects on the natural rate of unemployment” is, again, professor-speak for supply side effects.
I hope that cleared things up for you.
FWIW, I don’t think the R&R paper was particularly well written. I work for the government…worse yet, the Army, so I’m used to bad writing. I get plenty of practice every day; but the R&R paper is some pretty bad prose. I realize it was just a draft version, but I have a gated copy of the final version and it’s really not any better. One of my brothers is an English professor and he’d be banging his head against the wall if he had to read that kind of prose.
Mark,
If it is possible I would love to read your dissertation. Two of the most important works on the Great Depression I have ever read were dissertations. The candidates both took positions out of the mainstream but had more sound reasoning than anything I have seen written on the period. Let me know if it is possible.
Mark,
Let me give you an interesting name in academia and business, Professor Reuven Brenner.
2sb-
That may have convinced you, but it didn’t convince me. A lot of hand waving.
Rich Berger Okay, then let’s hear your version. Please explain what you think the Romers meant by “the economy’s normal, flexible-price level of output.” Please explain your understanding of “gap between actual and normal output.” Then tells us which curve is associated with “persistent departures of output from its flexible-price level.” Do you know what the Romers are talking about when they use the term “flexible-price level?” Finally, how do you interpret the term “the natural rate of unemployment.” Which curve is all about the “natural rate of unemployment?” I look forward to reading your reply.
Slug,
You do a great job of demonstrating that when the Romers talk about “supply side” they are actually talking about demand. I have found few demand side economists who understand supply side because theue entire methodology is wrapped up in consumption statistics.
This inability is usualy wrapped up in a lack of understanding of JB Say. He never said as Keynes and his followers loved to state that “supply creates its own demand.” This argument relied on reductive fallacy which was then turned on its head. If you do not understand the concept “one produces to consume” you are forever lost in a Bizarro World of effect preceeding cause, consumption preceeding production.
Ricardo Nice try, but a red herring. We’re not talking about your understanding of Say’s Law. We weren’t discussing whether or not the Romers and Keynes correctly understood Say. The question on the table is whether or not you correctly understood the Romers’ paper.
Sure-
From the last sentence of the section you referenced:
“Thus, the behavior of inflation and unemployment are exactly what one would expect if the
estimated long-lasting output effects of tax changes represented persistent departures of output from its flexible-price level. The evidence is not definitive, however: the inflation response is not precisely estimated, and it is conceivable that tax changes have large effects on the natural rate of unemployment.”
Awful lot of hedging there, wouldn’t you agree? If they wanted to say that the effects of exogenous tax changes were not due to supply-side effects, why didn’t they just say so?
I think the idea that you can identify a level of output that is optimal or natural is suspect. In a sense, this is an assumption of Keynesian economics that serves a purpose similar to that of the ether pre-relativity. It cannot be observed, only estimated by models. And if the effects of exogenous tax changes are as persistent as they observe, isn’t the attribution to a non-supply side factor a distinction without a difference?
Even their attempt to characterize the effects of exo changes as deviations from the flexible-price output level is very convoluted, using Phillips curve methodology, and even so they say that “The one important caveat to this discussion is that the response of inflation to the tax cut is not precisely estimated. For example, there is only one quarter for which the null hypothesis of no effect can be rejected.”
They further say that a tax increase is followed by a large and significant increase in unemployment – certainly not what the supply side might expect, eh?
The more I read the more I think that the entire edifice of modern macroeconomic theory is rotten, with assumptions piled on assumptions, and complex mathematics employed to produce what? The economy, like the weather and climate, is too complex to allow detailed and precise prediction.
Anonymous / Ricardo If you disagree with the paper, then fine. I have plenty of problems with it as well. But remember, you’re the one who brought it up. You’re the one who cited it as somehow supporting your viewpoint. I think we can wrap-up this whole long discussion with a simple summary: Ricardo thought that the Romer & Romer paper supported his viewpoint because one of the findings was that tax increases can sometimes hurt GDP growth and by extension tax cuts can sometimes help GDP growth. But upon further reflection Ricardo now believes that he was originally mistaken in his initial assessment of the R&R paper and that the authors were not saying what he thought they were saying. The poster Ricardo would therefore like to withdraw his previous statements and strike all references to the R&R paper from the record. Does that pretty well sum up where we’re at?
No.
Thanks Rich, 😉 Slug is a hoot sometimes.
Ricardo-
Agreed. I am sure there will be a future post where the fur flies again. I think this one has run its course.