October 18, 2007
The Debt-to-GDP Outlook in Plausible Scenarios
In updating a graph of the projected debt-to-GDP ratio, I was only slightly surprised to see that the out-years (still) look pretty grim.
Figure 1: Publicly held Federal debt to GDP ratio, by fiscal year (thick blue line); and baseline projection (blue); extend EGTRRA, JGTRRA, other expiring tax provision, and AMT indexing (red); and extend tax cuts, discretionary spending grows with nominal GDP, and reduce Iraq troop levels to 75,000 by 2013 (green). Source: CBO, The Budget and Economic Outlook: An Update August 23, 2007, and author's calculations.
To reiterate, the baseline is calculated under current law, and assumes the tax cuts expire, starting in FY2008. But some observers have argued that extending the tax cuts need not result in rising debt-to-GDP ratios. These figures show that's pretty implausible. More likely, given historical trends, is an increasing debt ratio, even if we reduce troop levels in Iraq to 75,000 by 2013.
Should one believe that FY2008 real GDP should grow slower than the August 23 projection (2.9% year-on-year), well then, the short term looks a bit worse under any scenario.
This is being submitted to the Facing Up Blog Carnival on the $9 trillion debt
Posted by Menzie Chinn at October 18, 2007 06:25 AMdigg this | reddit
Listed below are links to weblogs that reference The Debt-to-GDP Outlook in Plausible Scenarios:
» October 18, 2007 from PrefBlog
ABCP is in the news nowadays - and the Fed reports that outstandings are down another $11-billion over the week, as the unwinding / delevering continues. The total outstanding is now down about 21% from the July month-end figure. The WSJ has published ... [Read More]
Tracked on October 18, 2007 06:04 PM
What is the manner of AMT indexing for above? Equivalent to the one-year patch proposal currently being discussed?
Posted by: RD at October 18, 2007 06:50 AM
is that what is suppose to happen in slow growth environment per Keynes?
Posted by: s at October 18, 2007 08:00 AM
What increases GDP? I understand the CBO use of a baseline and hopefully no one here believes that it has any relation at all to reality. There are many difficulties attempting to forecast such a number. For example, not all tax cuts are equal.
In the first Bush tax cut it was largly a Keynesian consumption tax cut. Such a tax cut becomes distributed to inflation, conspicuous consumption, personal debt reduction, and some small amount to investment. The second Bush tax cut was much more Supply Side in that it reduced taxes for investment in capital used in production through more favorable depreciation treatment and capital gains.
Notice that the first tax cut was more of a one time event with little long lasting effects, but the second tax cut took advantage of the miracle of compounding in that the gains from production were company profits that allowed the company additional investment to generate in more gains in production. The investment in production is normally not realized for one to two years after the Supply Side tax cut but once they begin to generate gainst these are compounded especially if the tax cuts continue.
But reversing the tax cuts end the compounding in stages. First the immediate reduction in investment then in reducing the gains available for reinvestment.
And this is not even considering tax increases from state and local government and the AMT that are a disincentive with negative compounding.
Similarly it is difficult to judge the spending on military actions. If a reduction in military action allows attacks that disrupt domestic production there can be a combination of redeploying the military deterent as well as increasing spending on national security. Perhaps even more of a concern is the reduction in civil liberties that arise from out unwillingness to deal with known threats.
But academically the graphs are amusing.
Posted by: DickF at October 18, 2007 08:11 AM
There was talk today of Capitol Hill to end the AMT without any corresponding increase in other revenue or decrease in spending. This was evidently the first time that the subject was floated seriously, which was why it was reported on CNBC.
That would "cost" the treasury $800B over 10 years.
Not phasing out the Bush tax cuts supposedly "cost" $4T over 10 years.
I'm kind of surprised that the AMT generates so little revenue.
Posted by: Buzzcut at October 18, 2007 10:31 AM
I don't think anybody's called me pessimistic to my face.
And I probably wouldn't either.
You'll just have to invite all your "regulars" over to your Lake Geneva estate for drinks and we'll see.
Posted by: Buzzcut at October 18, 2007 10:37 AM
Yeah! And invite some of your newbies, too!
Posted by: CoRev at October 18, 2007 10:52 AM
I think CBO is still way too optimistic.
We are accumulating infrastructure deficits and social deficits that impact the economy. We haven't dealt with exploding medical care costs as they impact Medicare A and D, not to mention VA, Medicaid and SCHIP. I also doubt that CBO has taken into account the effects of dollar readjustment, which will lead to slower growth, higher inflation, and a weakening of the productive base of the economy. Even the IMF has only gradually been waking up to the magnitude of the imbalances.
It's more than grim. It's catastrophic.
Posted by: Charles at October 18, 2007 11:35 AM
RD: The AMT fix involves indexing for inflation. See page 19 of the CBO update.
s: Not sure I understand the question. Certainly tax revenues tend to fall and transfer payments tend to rise during economic slowdowns. This is why I often refer to the cyclically adjusted budget balance as an indicator of fiscal stance.
DickF: We've had this debate in previous exchanges on this subject. First, CBO uses a variety of models at the short and longer horizons. Second, I think these models would be well accepted as reasonable in mainstream macroeconomic discussions. Third, I am hard pressed to think of a different model -- accepted by mainstream economists -- that would yield drastically different trajectories, given best guesses about demographics, population growth, capital accumulation, and trend total factor productivity growth. While you may believe these estimates have no relation to reality, at least CBO has attempted to show the degree of uncertainty associated with its forecasts, expressed as a function of its track record in terms of forecast errors. See the fan diagram on page 22, as well as this technical document.
Buzzcut and Corev: OK, when I get that estate, you'll get the first invitations (by the way, I had to look up where Lake Geneva was).
Charles: On the forecast/projection front, I agree that CBO is (now) probably too optimistic, but their forecast -- probably completed more than two months ago -- was likely in the middle of private sector and professional forecasts available at the time. Regarding the budget projections, remember CBO is charged not with projecting what is likely to happen, but what will happen under current law. CBO, in a series of very easy-to-use spreadsheets, has provided the means to examine what happens if various provisions (which you and I might disagree upon regarding plausibility) are implemented.
Posted by: Menzie Chinn at October 18, 2007 12:15 PM
Menzie, you're obviously working too hard if you don't know where Lake Geneva is! Come on, it's where the elite of Madison and Chicago meet.
Posted by: Buzzcut at October 18, 2007 12:42 PM
Oh, and the 10 year rate today is 4.5%.
Rates are powerfully influenced by the overall economy. They are weakly influenced by debt levels.
Posted by: Buzzcut at October 18, 2007 12:57 PM
I understand about the CBO. Could you graph actual numbers against CBO numbers and your numbers. I think that we will find that actual numbers will be significantly different because there are simply too many variables.
My primary reason for continuing to make my point is that congress and its many agencies make decisions based on such questionable information resulting in ruined lives. The Democrats actually believe that increased taxes will bring in more revenue and allow them to balance the budget, but we are on the wrong side of the Laffer Cruve.
When the government stops intervening and ruining lives I will stop criticizing their data.
Posted by: DickF at October 18, 2007 01:48 PM
I recall at the end of the Clinton years the CBO projected the national debt would be paid off in about 15 years due to the surpluses. So much for CBO projections.
In our modern era of fiat money with floating exchange rates debt and deficits are just accounting issues. The government taxes and sells treasury securities to drain reserve balances created by government spending in order to maintain the Feds target Fed funds rate. The real issues are maintaining the value of the currency (preventing inflation) and the economic impact and social benefit of spending and taxing. Looking at the graph of debt/gdp ratio and growth rates and interest rates and you will find there is very little correlation. As a matter of fact, growth tends to slow when the debt/GDP ratio drops. Am I invited for drinks too!
Posted by: MarkG at October 19, 2007 06:29 AM
MarkG: I think you forget about two tax cuts; the CBO projections are based on current law. See the comments on the previous post, and reference to the January 2002 CBO update for the impact of the first.
Posted by: Menzie Chinn at October 19, 2007 07:14 AM
Menzie: My comment was not on the validity of the CBO projection, just the usefulness. What is the economic consequence of varying debt/GDP ratios and is there any historical data to support your answer?
Posted by: MarkG at October 19, 2007 07:27 AM
1) I have a dumb question. Are your projections and the CBO baseline all based on dynamic scoring or static?
2) Also, I'm most interested in what happens to our fiscal situation a couple of decades out (2020, 2030, 2040) as entitlement costs -- per current eligibility and benefit structure -- soar. Are you able to project over a longer period, perhaps with ranges due to the greater uncertainty.
My Holy Grail, by the way, would be to find a set of graphs that show impact on GDP and on debt-to-GDP of various scenarios of taxation (some or all of the Bush tax cuts extended vs. expiring) and spending (specific cuts in projected entitlement benefits and eligibility, as well as cuts in projected discretionary spending). That would really provide policy-makers and voters important and useful insight into the degree of overall sacrifice necessary to get our long-term fiscal house in order and the trade-offs involved, so we can debate and set responsible fiscal policy with sacrifices that reflect our priorities. Can anyone direct me to such information? If so, please comment here or email me at Brooksbud@aol.com
Posted by: Brooks at October 19, 2007 07:40 AM
Menzie says, "Regarding the budget projections, remember CBO is charged not with projecting what is likely to happen, but what will happen under current law."
True, Menzie, and they can't be expected to foresee wars and natural catastrophes.
What happened in the wake of the 1994 elections was despicable. Alice Rivlin had done a superb job of prediction, so she was forced out of CBO and into OMB, where she continued to do a superb job of prediction. Meanwhile, the Gingrich Congress started politicizing the CBO. Their record was of prediction was very poor in the 1990s. I think it's improved, but it's still being bent to ideological winds.
I really don't care that much about ideology, but when professions like economics and science become Lysenko-ized, it represents an extreme form of moral rot on the part of leadership. Punk supplysideism is purely ideological. It would have long ago died were not huge amounts of cash and media pumped into maintaining the vampire lies.
Posted by: Charles at October 19, 2007 08:15 AM
I am not quite as cynical as DickF, but I don't place a lot of trust in long-term projections. I do believe that people put far too much trust in projections - that 5.6 trillion surplus was commonly considered to be "real". Consequently, when new projections showed a sharply lower number, President Bush was blamed for squandering it. I read the 2002 CBO report you cited it appears that the surplus was reduced by one trillion on account of changed economic circumstances and "technical" factors. What a difference a year or two made.
After a year or two, so much uncertainty enters in that the reliability of the forecase declines sharply.
Is any forecast better than none? I am a consulting actuary specializing in pension plans. I have done many projections, and my conclusion is that you can get into a lot of trouble with single scenario deterministic projections. What is much more useful, although harder to understand, is a range of projections, under alternative scenarios. A range gives a sense of the potential variability and can help an audience understand what can happen.
I have no reason to impugn the forecasters at the CBO and the OMB. I am sure that they are smart, honest and diligent professionals. The limitations are in the nature of projections.
Posted by: Rich Berger at October 19, 2007 09:22 AM
One curious trend in the graph is that the ratio seems to have peaked and the trend is down yet this period is close to the time the Iraq conflict began. Does it seem unusual to anyone that such huge military spending as reflected by the green line actually had virtually no effect on the ratio? Another curious thought is that this peak seems to be about the time the second Bush tax cuts became effective. Hmmmmm?!
Posted by: DickF at October 19, 2007 10:29 AM
From a cynic - very good post!!!
Posted by: DickF at October 19, 2007 10:30 AM
You are on to something. Everyone should notice, that with the exception of the Mid-1980s, the spikes in the debt to GDP ratio occur around recessions. If the growth of Nominal GDP exceeds the deficit as a percentage of GDP, then the ratio will fall. The reverse happens when there is a recession: the growth of Nominal GDP stalls and deficits rise as tax receipts fall with economic activity. Anyway, the alternative projections shown aren't that big a deal. Those scenarios would still put the US at one of the lowest levels of debt to GDP among major industrialized nations.
Posted by: Tyler at October 19, 2007 10:52 AM
We can't just look at our debt-to-GDP in isolation and conclude it's not a problem. We need to look at our debt-to-GDP in the context of anticipated expenses -- meaning our enormous unfunded entitlement liabilities -- as well as the projected decline in the worker-to-retiree ratio over the next couple of decades. To draw an analogy, there's a big difference between a couple (two workers) who are parents of one young child having a given debt-to-income ratio vs. a single parent with that same debt-to-income ratio who has three kids nearing college age.
We need to get our debt-to-GDP down, rather than increase it, in anticipation of these predictable entitlement expenses, which will be huge due to the number of retirees even if we cut benefits and eligibility as much as is politically feasible.
Watch video here:
Posted by: Brooks at October 19, 2007 11:08 AM
In updating a graph of the projected debt-to-GDP ratio, I was only slightly surprised to see that the out-years (still) look pretty grim.
Though on the plus side, it does look significantly better than before the update. Am I correct that the red line now is the same scenario as the green line in the previous graph? It seems so, since both are the "extend EGTRRA, JGTRRA, other expiring tax provision, and AMT indexing" graphs. The CBO baseline is also quite a bit better as well, even if unlikely.
It's another indication of how much economic growth (and assumptions of such) affect these projections. Even the actual one year performance was different than the one year projection last year, with a small decline instead of a small increase.
Posted by: John Thacker at October 19, 2007 11:38 AM
The difference between the 37.3% debt to GDP ratio expected for FY07 in the August '06 CBO report and the 36.4% result that we're now getting is still substantial. That's actually real, but since the CBO projections always use a baseline that's known to be unrealistic (esp. as regards current economic growth), the future projections are always problematic, none moreso than when an Administration starts bragging about future surpluses when we're at the height of the economic cycle.
Posted by: John Thacker at October 19, 2007 11:47 AM
And of course the 2006 projections were particularly ridiculous since the CBO's baseline predictions were required to claim that the Katrina supplemental spending would be repeated and expanded every year. A lot of the improvements in the projections came from that going away.
Posted by: John Thacker at October 19, 2007 11:49 AM
How do we know when we're "at the height of the economic cycle"?
As for the Bush Administration bragging about deficits coming down, my problem with that is that it's been happening DESPITE their policies, not BECAUSE of them. Spending has gone up substantially (even aside from war / homeland security) and revenues are much lower than they would have been without the Bush tax cuts.
Posted by: Brooks at October 19, 2007 11:53 AM
Brooks, you've inadvertently stepped into one of my hot button mine fields.
Spending has gone up substantially (even aside from war / homeland security) and revenues are much lower than they would have been without the Bush tax cuts.That mine field? The woulda/coulda/shoulda game. Revenues are much lower??? Maybe? Care to show a proof of that? See what I mean about the woulda/coulda/shoulda game.
Now, your comment about spending is undoubtedly the absolute truth, with one little exception. Spending growth has been held substantially below the growth of revenue. The deficit is in retreat because of it.
Are you proposing that spending growth be held at zero (below inflation) or even negative (way below inflation?) In my woulda/coulda/shoulda game that policy would devastate an already shaky economy.
Posted by: CoRev at October 19, 2007 12:42 PM
Glad to be in your minefield. My contention that the Bush tax cuts have had a substantial net negative impact on revenues is based partly on my own sense of likelihood based on Laffer Curve implications, but more so on the fact that there is a strong, broad consensus (with very few exceptions) on that point among well-credentialled economists -- including conservative economists and Bush's own current and former top economists. I've researched it fairly thoroughly, and the consensus is clear.
As for the Laffer Curve implications to which I referred, here's an illustration: For simplicity, let's assume a flat tax scenario. A 5-point tax cut from 90% to 85% requires an expansion of the tax base (from incremental GDP and reduced tax avoidance) of only 5.9% to be revenue-neutral, while increasing the incentive to work and invest by 50% (from 10 cents per dollar earned to 15 cents), in addition to providing some demand-side stimulus. Compare that with a 5-point tax cut from 25% to 20%: the required expansion of the tax base is 25% while the incentive only increases 6.7% !! Far, far, far less plausible. That's the logic behind the Laffer Curve, and in my view it makes the consensus among economists about the Bush tax cuts quite intuitive, although I'm sure their conclusions are based on more sophisticated theory and analysis.
You can always say "woulda, coulda, shoulda" about just about anything, and economics is obviously an inexact science, but that doesn't mean that economists can't reasonably draw the conclusion that a particular tax cut is highly unlikely to have "paid for itself".
Posted by: Brooks at October 19, 2007 01:05 PM
Also CoRev, IF you were correct (which you are not) that we simply have no idea what the net impact on revenues of the Bush tax cuts has been, then I assume you agree that the Bush Administration is wrong to claim credit for the increases in revenues over the past few years, right?
As for spending, the fact that revenues have been increasing faster than spending hardly lets Bush (or Congress) off the hook. First of all, it's not at all clear to me that spending was increased in anticipation of higher revenues, as opposed to independent of projected revenues (and in addition to increased non-Defense discretionary spending, what about that fiscally reckless expansion of Medicare [Part D] in 2003?). Second, so what? Based on our long-term fiscal outlook (and particularly unfunded entitlement liabilities) the level of spending, particularly relative to the level of taxation, has been highly irresponsible, period.
As for my policy prescriptions regarding spending, I don't claim to have very specific proposals, but in general I think we need to begin phasing in a higher retirement age, greater means-testing of entitlement benefits, reduced Social Security benefits (partly by indexing Social Security to price inflation rather than wages), and reduced Medicare benefits (perhaps phasing in elimination of Part D, at least on an aggressive means-tested basis). Obviously we should also be vigilant on pork in the discretionary budget, but that's small potatos compared to projected entitlement costs. And while I don't have a clue how to increase efficiency in healthcare, that's critical as well.
As for the effects of lowering spending vs. baseline projections and/or vs. current spending and its impact on the economy, first of all I think we're better off letting the Fed manage economic cycles (they can have more timely impact and their decisions are driven more by economics than politics), but moreover, if your concerned that cutting spending (or raising taxes) now will harm our economy and standard of living, you should be MUCH more concerned about what will happen if we don't take such actions to address our long-term fiscal imbalance and instead just wait for the feces to hit the fan.
Posted by: Brooks at October 19, 2007 01:31 PM
There are two sides to this ratio. There is the debt side and there is the GDP side. You can change the ratio by decreasing debt or by increasing GDP. While I agree that the debt side is excessive, every debt dollar has a constituency to fight against its cut, but everyone is a constituent of increasing GDP because it benefits everyone. So, why not implement policies that allow the economy to increase GDP rather than reduce it?
I am sure you know the miracle of compounding. Now when you save for future consumption does your saving "pay for itself?" Usually the answer is no, but what you do is reduce consumption to allow for the saving for a period of time because you anticipate a gain at some point. Now think about this in relation to international growth. There are many countries, the Balkans, China, and others, who have reduced taxes, postponed consumption, to gain an incremental increase in their national growth rate. This is one reason the US grew faster than the UK and why today other countries are growing faster than the US.
Now there are two reasons that you save. As I stated one is future consumption but another is to accumulate wealth. By saving in an interest bearing instrument you can use the miracle of compounding by reinvesting your interest so that interest is paid on interest.
Supply Side tax cuts increase production. Now the increased production may not offset the reduction in tax revenue right away but it has increased production. None of the studies deny this. Not the tax cuts act like the interest rate and the increased production reinvested takes advantage of compounding.
Now let's reflect back to our interest example. If your instrument stops paying interest what happens to the miracle of compounding? Now consider this. Assume that the fees you pay to maintain your saving instrument are also increased? Can you begin to see the effects or tax cuts and tax increases?
It appears that you look at the Laffer Curve primarily from the perspective of increasing the incentive to work harder, but this is not the primary contributor to the effects of the Laffer Curve. The primary contributors to the positive Laffer Curve effect are the exponential increase in production because of compounding and a reduction in non-compliance with tax laws, meaning a 100% tax rate does not end production, but it pushes production into the black market. Tax revenue is lowered significantly by producers engaging in barter, or producing “off the books,” or changing investment instruments to take advantage of tax breaks (municipals for example). When tax rates are reduced on the rich they pay a greater percentage of total taxes because they shelter less of their income.
Finally, let me ask you if tax increases pay for themselves?
There is much more on this but I have gone on for too long. Hopefully I have stimulated thinking concerning the debt to GDP ratio.
Posted by: DickF at October 19, 2007 02:10 PM
Your argument, if I'm understanding it correctly, is based on flawed analysis.
First, you seem to equate tax cuts with increased savings/investment. Apparently you are overlooking the fact that much of the incremental after-tax income would go to consumption rather than saving/investment. Granted, less so for a cap gains tax cut, but still the case.
Second, you overlook the fact that, since we are carrying a publicly-held national debt (as well as running annual deficits) any additional dollar that is returned to the taxpayers rather than used to reduce the amount we are adding to our debt (or to pay down debt) is a reduction of national savings on the part of the Federal Government, even if taxpayers will partly compensate for that reduction by saving and investing the extra after-tax earnings (to the extent that they don't use it for consumption). Additionally, this impact on our debt level causes us to incur more interest expense, due to the higher debt level itself and insofar as higher debt creates upward pressure on interest rates, ceteris paribus.
Which brings me to your emphasis on compounding. You should be applying the time value of money to all the components of this picture rather than only to private savings. Government has a cost of capital (as well as an opportunity cost), including, but not limited to, it's interest expense, so any NPV analysis we do here must apply compounding to the revenue lost due to the tax cut until the point in some future year at which a tax cut today MAY (arguendo) ultimately have a net positive impact on nominal (or real) revenues, and a discount rate must be applied to any incremental revenues gained in those out years (substantially diminishing their present value).
As a note, you mischaracterized my perspective on the Laffer Curve. Perhaps you missed it or misunderstood, but I referred to "expansion of the tax base (from incremental GDP and reduced tax avoidance)". The "tax avoidance" is not just people cheating on their taxes more, but rather includes black market activity, barter, etc. Additionally, I described (accurately) the other major component of the logic of the Laffer Curve, which is the simple algebra of break-even analysis.
Re: your question, "let me ask you if tax increases pay for themselves?", at current tax rates (or anything anywhere close to them, as opposed to theoretically extremely high tax rates), the answer from almost all economists is "yes" (I presume, since your question is, for the most part, merely the converse of the tax cut question), and that consensus seems sensible to me for reasons I've explained in a previous comment, so I feel pretty comfortable with considering it at least much more likely to be valid than not.
Posted by: Brooks at October 19, 2007 03:03 PM
Brooks, I'm sorry but to me, economic discussions too often "based partly on my own sense of likelihood based ..." and then the writer fills in the blank. But, when we are into consensus science:
I've researched it fairly thoroughly, and the consensus is clear.Just isn't science, not even an inexact science. BTW, if I remember correctly, you are the only one claiming that the tax cuts pay for themselves. Not something most fiscal conservatives have claimed in quite some time. As to the rest of your 0105 Pm comment it is mostly more of the that woulda... impossible to prove.
In you later comment this is just wrong:
then I assume you agree that the Bush Administration is wrong to claim credit for the increases in revenues over the past few years, right?If we don't play the woulda... game and use real numbers then absolutely, Bush can claim his revenues rose. They are definitely higher than they were during the recession, which above all is what his policies were trying to accomplish. Are they higher than .... you fill in the blank? I don't know unless your comparison is a real world example.
And then finally, we get to this comment:
First of all, it's not at all clear to me that spending was increased in anticipation of higher revenues,...it appears you forget the subject of this article. Short term/annual projections seem pretty accurate. To think that budgets are not prepared with them as inputs just seems, naive.
Posted by: CoRev at October 19, 2007 04:58 PM
You're quite mistaken pretty much across the board on the points in your (3:03pm) comment.
You say I can't argue a point of view based on what makes sense to me (based on the dynamics and relative magnitudes of forces involved, as I've explained clearly) nor on what the expert consensus is. Well, you'll have to forgive me if I disagree with your unusual requirements for a rational basis for a conclusion.
Where and why you think I have claimed that tax cuts pay for themselves I have no idea. And you are quite wrong that such a claim is non-existant among self-proclaimed fiscal conservatives (Just look at today's National Review, for one thing, if this is recent enough for you http://article.nationalreview.com/?q=YzRlYWM0ZjE1MjlkYTExYmQ2MDMxYzEwOTQ1MGEyYzQ= ).
As for your "woulda, coulda" argument, I've already explained why that quite weak, to say the least.
As for my question re: Bush taking CREDIT for the the increases in tax revenues, instead of answering my question you erected your own straw man (that revenues are higher) and responded to that. Was that deliberate, or do you not understand the difference between a positive devlopment and one rightfully claiming credit for that development (i.e., having caused or at least contributed to it)?
As for your last comment regarding spending, you make an argument that you think refutes my assertion, but it doesn't. I'm saying it's quite possible that the spending increases were decided upon independent of projected revenues, rather than based in part on those projected revenues. And one big example is the major expansion of Medicare (Part D) in 2003 -- I repeat, 2003 -- prior to the increases in revenues, and certainly not based on long-term projections of increased increased revenues to match the long-term incremental liability incurred.
So pretty much every single point in your comment was erroneous in one way or another. Seems like the "minefield" about which you boast is full of duds.
Posted by: Brooks at October 19, 2007 06:10 PM
Just for kicks, if the consensus view among economists, or if just my view, were that a tax cut from our current system to a flat 1% tax on all forms of income would have a net NEGATIVE impact on revenues, would you say "We just have no way of knowing, woulda/coulda/shoulda"?
How about a 5% flat tax? 10%?
Would you just throw your hands up in the air and say we can't possibly have a good idea of the probability that such a tax cut would have a net negative impact on revenues, or would you think it through in some way, and if so, how? (hint: I've already given you the bulk of the answer in a previous comment)
Posted by: Brooks at October 19, 2007 06:27 PM
Brooks, if it isn't obvious, my penchant is to use real numbers. Yes, we could calculate anticipated revenue from your proposed flat taxes. I would guess they could be estimated +/- 5% for the first year without stretching too many assumptions. I do not know if that is accurate enough to budget from.
I base my estimate on the knowledge that much of the newly taxed income is not tracked in today's tax system. For example, it makes me think of the tax category for gambling winnings and losses. Investment returns for a large number of returns are not reported, or reported inaccurately due to today's deductions structure. Add to that the 15M+ of not reported returns, and the 43% of current returns that do not pay taxes, which to me, after realizing they now owe, would increase the number of non-reporters.
Posted by: CoRev at October 20, 2007 08:13 AM
I have tried to be concise but you gave me a lot to respond to so here goes.
I was not equating tax cuts with increased savings/investment but tax rates with investment interest rates, payments.
Many critics of Supply Side tax cuts make a mistake assuming cuts go to consumption rather than saving/investment because of Keynesian analysis. Again, note the distinction between the first Bush tax cut and the second. The first, Keynesian not a Supply Side went to consumption.
Your analysis of federal government debt is incomplete (Economics in One Lesson). Each dollar taken from taxpayers to reduce the federal debt is a reduction in private sector investment and by extension production. Federal debt is no problem as long as it can be serviced, just like there is nothing wrong with financing a house if you can pay the bill. Government debt does not finance production, increase wealth, while private debt does. A surplus, tax revenue in excess of federal debt payment, should be returned to the productive economy to generate additional wealth increasing government revenue. Three is minimal return accelerating government debt retirement, while there is great loss extinguishing productive resources.
Your critique of compounding assumes simple savings rather than production of wealth in the private sector while ignoring that government debt consumes rather than producing. Government revenue comes from private sector production not from government spending. If this were not true there would be no need for taxes.
Sorry if I misunderstood your point about the Laffer Curve. Econometrics divorced from behavior can be misleading. For example with a reduction in tax rate from 20% to 25% a tax payer can realize this 20% reduction in taxes by a simple change in investment mix the incentive is nearly 100% with little effort. Anyone who has run a business, especially a grocery store, realizes the gain from even a 1% addition return. This is obvious looking at the increased percentage of taxes paid by the rich with a small decrease in tax rate.
Your answer to my question “does a tax increase pay for itself” did not demonstrate any net increase in wealth to pay for removing the resources from the productive economy. A 25% tax increase (from 20% to 25%) taken to reduce government debt will at best save the interest rate of the debt but it extinguishes wealth from production. It is impossible for a tax cut to pay for itself. Once again if it did there would be no need for taxes.
You must understand that government is like the cost of a security guard; it is a net cost to production and wealth creation that can only be justified by reducing loss. If the cost of government does not reduce loss, there is a net loss of wealth.
Now you may justify the loss of wealth for social reasons but not through economics. The loss is a loss even if justified. That is a political decision.
Posted by: DickF at October 20, 2007 08:19 AM
The little digs aside, I am impressed with the general tenor of this discussion/debate. Civil, with good points made by the discussants.
I favor tax cuts for several reasons. First, I prefer to keep more of my income. The biggest infringement on my liberty is taking the fruit of my labors. I do not think the government spends my money carefully. Although Menzie periodically highlights Iraq war spending, I don't see much analysis of the rest of the budget for effectiveness.
Second, I think Laffers's analysis is conceptually correct. The empirical question is: what is the optimal level of taxation for revenue generation? Reduced tax rates have two stimulative effects: they increase the return to effort and they reduce the value of tax shelters. For the 39% tax bracket (before the Bush cuts), an income earner took home 61% after Federal taxes. The 10% tax cut increased the take to 35%, and the after tax to 65%, for an increase of almost 7%. Tax shelters become 10% less valuable.
Finally, reductions in tax rates reward effort and investment, which raise production. I think the emphasis on consumption is wrong; societies grow richer not by consuming their product, but by investing and working harder and smarter.
I understand that there are studies showing that tax cuts do not have much affect, and other studies showing that they do have a substantial effect. I do think the question is not easy to solve - you can't rerun history with and without tax cuts. I tried to reconstruct expected versus actual revenues with budget forecasts and hit a dead end (couldn't find the time to pursue it further).
Posted by: Rich Berger at October 21, 2007 11:08 AM
I understand that there are studies showing that tax cuts do not have much affect, and other studies showing that they do have a substantial effect. I do think the question is not easy to solve - you can't rerun history with and without tax cuts. I tried to reconstruct expected versus actual revenues with budget forecasts and hit a dead end (couldn't find the time to pursue it further).That's why I do not think too highly of the strawman comparisons we see on econo-blogs all the time. I too think it can be done, but by the time a study is completed that is relicatable AND consistent, then an econo-model of extraordinary proportions and usefulness is available to all. Right!
For our next trick we will model the Global climate, so we can determine the effects of ..., ..., ..., and on, and on, and on, until we can determine how a single variable actually effects the climate.
Posted by: CoRev at October 21, 2007 03:42 PM
Re: “Many critics of Supply Side tax cuts make a mistake assuming cuts go to consumption rather than saving/investment”
I don’t know if “many critics” make that assumption, nor do I consider myself a critic of supply-side tax cuts, only of the invalid rationale that some offer that they will have a net positive impact on revenues. I’m fully aware that some types of tax cuts, such as the 2003 cuts on cap gains and dividends, stimulate incremental investment, and tax cuts on labor income can also increase the supply of labor, both of which have obvious benefits, ceteris paribus. I also assume that the long-term revenue feedback effects of the 2003 tax cuts will be greater than those of the 2001 tax cuts, but even the former has not and will not get near 100% (i.e., self-financing, “paying for itself”, so to speak), let alone have a net positive impact on revenues, according to the consensus view of economists.
Re: “Federal debt is no problem as long as it can be serviced, just like there is nothing wrong with financing a house if you can pay the bill.”
True, there’s nothing inherently bad about Federal debt per se (few would argue with that), but we have enormous unfunded entitlement liabilities and a projected deterioration in the worker-to-retiree ratio, which, coupled with our current debt-to-GDP ratio, means that we are on an unsustainable fiscal course that will blow up in a couple of decades on it’s current track. Economists from left to right and in between all agree on that point. Presumably the Federal Government will not default on its debt obligations, so it will always be able to “pay the bill”, but the longer we dig ourselves deeper into this fiscal hole, the greater the ultimate pain will be in terms of both spending cuts and taxation.
Regarding your other comments regarding what you think is my view of government debt, I think you’re misunderstanding my point. I’m not suggesting that, in all situations, retiring another dollar of government debt is necessarily a better choice than returning it to taxpayers. And in general, I think the private sector is a much more efficient allocator of resources, and in many cases a dollar invested in the private sector will yield a greater return than the Federal Government’s cost of capital (or some other appropriate discount rate to calculate present value), albeit with risk associated with that ROI vs. the (near) certainty of the Federal Governments requirement to pay interest on the additional debt along with principal.
My other points regarding flaws in your approach still hold. (1) You don’t seem to consider the large portion of incremental after-tax income that would be used for consumption rather than savings or investment, (2) you seem to apply the principles of compounding and, more generally, the time value of money only to incremental growth due to private sector investment resulting from lower taxes rather than applying these principles to the additional debt as well (see my previous comments for elaboration).
Re: “A surplus, tax revenue in excess of federal debt payment, should be returned to the productive economy to generate additional wealth increasing government revenue. Three is minimal return accelerating government debt retirement, while there is great loss extinguishing productive resources.”
Do you really mean that?? Do you realize what you just said? You are saying that the government should make it’s payments on the debt and nothing else, meaning that our ENIRE budget (other than payments on the debt) should be deficit-financed. I hope that was a mistake, because otherwise it’s insane. Perhaps you just meant a surplus as in revenue in excess of spending, which would make more sense. Is that what you meant?
Re: “Your critique of compounding assumes simple savings rather than production of wealth in the private sector while ignoring that government debt consumes rather than producing.”
I don’t know why you are attributing such an assumption to my argument. I assume taxpayers would use incremental after-tax income after a tax cut for various combinations of consumption and savings & investments of all sorts.
Re: “Government revenue comes from private sector production not from government spending. If this were not true there would be no need for taxes.”
What does that mean?? Who thinks government revenue comes from government spending? I mean, perhaps you’re trying to represent the Keynesian concept of using fiscal policy to manage economic cycles – e.g., cutting taxes and increasing spending to add to aggregate demand during a recession – but it doesn’t make sense to say that government revenue comes from government spending, so who says that??
Re: “Your answer to my question “does a tax increase pay for itself” did not demonstrate any net increase in wealth to pay for removing the resources from the productive economy. A 25% tax increase (from 20% to 25%) taken to reduce government debt will at best save the interest rate of the debt but it extinguishes wealth from production. It is impossible for a tax cut to pay for itself.”
That conclusion does not follow. Let me ask you this: If we cut our tax rates on labor and investment income to a flat tax rate of 0.1%, would the long-term cumulative net affect on revenues (in present value) be positive or negative?
Re: “Now you may justify the loss of wealth for social reasons but not through economics. The loss is a loss even if justified. That is a political decision.”
I’m not introducing politics into this discussion, and I’m certainly not making a case for using tax policy for a greater degree of wealth redistribution. I’m making a simple point regarding the relationship between changes in tax rates from current levels (or from levels prior to the Bush tax cuts) and revenues. I am more concerned with debt-to-GDP than with debt level per se, and economic growth is obviously relevant to the former, but your argument seems not to make sense to me. Moreover, your view is at odds with the broad, strong consensus view of economists on this question. I invite and encourage you to visit my posts here http://logicizer.townhall.com/g/c5ecb3cf-2712-4f5a-ad89-7ae03da99280 and here http://logicizer.townhall.com/g/f48d2bf3-1c51-4592-aa46-191f089d752f .
Posted by: Brooks at October 21, 2007 04:47 PM
It appears that we may agree more than we disagree, but you have still not addressed the issue of government being a cost rather than a producer. This is an important consideration in our discussion because too much of the rhetoric on tax policy and spending consists of the claim we are "investing" with tax revenue.
When I talked of servicing the debt I was including spending on government service. I thought that would be clear from my comments on government being necessary for general police actions.
My primary point is that increasing taxes to balance the budget is counter productive. We seem to agree that government obligations are the problem but this cannot be solved by pulling resources from production to generate government spending. Hoover already tried that.
Thanks for the links to your posts. I am sure I will enjoy them. I will let you know.
Posted by: DickF at October 22, 2007 05:48 AM
Re: "you have still not addressed the issue of government being a cost rather than a producer."
I can't tell if I'm just not understanding what you're trying to say or if you're saying something so obvious that it's silly to say, and if the latter, I think I did address it. Can you take another shot at explaining what you are talking about?
I hope you do check out those links. One provides strong support for my contention that there is a strong, broad consensus among economists that the Bush tax cuts have NOT contributed to higher revenues (and have had a NEGATIVE net effect), and the other illustrates why it is unlikely that tax cuts such as Bush's (across the board cuts of labor income rates along with the capital gains & dividends rate cuts, both from the particular pre-tax cut levels) would generate enough expansion of the tax base to be revenue-neutral.
Posted by: Brooks at October 22, 2007 04:26 PM
Your statement that "....have still not addressed the issue of government being a cost rather than a producer", is not alwasy true. It depends on what the government spends the money on.
Example: Schools, infrastructure, defense - as opposed to non-investment type of spending (loosly defined as pork barrel spending).
We have the same situation in private economies: Mortgaging your house to invest in something worthwhile is good, while doing it to increase your consumption of trivial stuff (beer and TV) is not.
In either case you are borrowing.
The problem isn't "the government is using up our monies (too large a portion of GDP)" the problem is how they are spending our monies.
Sometimes what constitutes wasteful spending is clear cut (e.g. Alaska bridge to nowhere), somethimes it depends on your political view (Iraq war), but most of the time most of us agree on what is and isn't useful (roads, schools etc.)
This is the problem with the Laffler curve, the "pays for itself-ness" of a tax cut depends on what the money is spent on (in addition to the tax rate before and after).
Posted by: bellanson at October 22, 2007 05:38 PM
Brooks, and bellanson, you folks keep arguing with yourselves :
...would generate enough expansion of the tax base to be revenue-neutral.I may not remember completely, but no one but you folks have claimed tax cuts pay for themselves.
Brooks, as for
I hope you do check out those links. One provides strong support for my contention that there is a strong, broad consensus among economists that the Bush tax cuts have NOT contributed to higher revenues (and have had a NEGATIVE net effect),...The article surely makes a case for the consensus view you state. The comment that caught my eye was one reason for tax cuts is to control spending.
Posted by: CoRev at October 22, 2007 07:42 PM
Re: "Brooks, and bellanson, you folks keep arguing with yourselves...I may not remember completely, but no one but you folks have claimed tax cuts pay for themselves."
I have no idea what you're saying, but you seem confused over what has been said by whom.
Re: controlling spending as a rationale for tax cuts, that's the "starve the beast" rationale -- cut taxes to lower revenues, putting pressure on the politicians to spend less. Whether or not it is effective is debated by economists, but you should note that it is incompatible with the contention that tax cuts increase revenues (obviously a given tax cut over a given time period cannot both reduce and increase revenues).
On the "starve the beast" theory, there is debate among economists and the jury is out among economists on whether or not lower taxes induce lower spending. It may seem intuitive that lower revenues would induce lower spending, but an intuitive argument is made the other way as well -- that borrowing to spend rather than taxing to spend shields taxpayers from the pain of spending and reduces the incentive to curb spending. In addition to this intuitive aspect, the empirical evidence is debated. A debate emerged among some economists after this piece by William A. Niskanen, Chairman of the Cato Institute and former member and acting chairman of President Reagan's Council of Economic Advisers. He holds a Ph.D. in economics from the (famously free-market) University of Chicago, which has honored him with a lifetime professional service award. http://www.cato.org/pubs/policy_report/v26n2/cpr-26n2-2.pdf . Niskanen concluded that tax cuts led to HIGHER, not lower spending. Now, supply-side economist Gregory Mankiw, who was Chairman of G.W. Bush’s Council of Economic Advisors and is an economics professor at Harvard, noted that another prominent economist reached the opposite conclusion, with some differences in methodology (with perhaps some advantages), albeit with less recent data vs. Niskanen's analysis. Mankiw concluded that the jury is still out on which is correct -- it's just not clear if "starve the beast works" (i.e., if tax cuts lead to lower spending). http://gregmankiw.blogspot.com/2006/06/starving-beast.html .
I have a couple of html questions, if anyone can please answer:
1) How do I make links "live" in a comment?
2) How do I set apart quotes of others (different font and/or indentation)?
Posted by: Brooks at October 22, 2007 08:27 PM
For quotes type blockquote .....then copy your quote...
then type /blockquote and surround both words with a left and right>. I have tried @ 6 ways to print the left version of this > and it doesn't show.
Posted by: CoRev at October 23, 2007 04:11 AM
Thanks, CoRev, but it doesn't seem to be working, at least not in Preview. If this doesn't work, what am I doing wrong?
Posted by: Brooks at October 23, 2007 09:05 AM
How do we know when we're "at the height of the economic cycle"?
Largely we don't, but even when there are signs the CBO (for understandable reasons) is forbidden from ignoring them; just as they're forbidden from treating the Katrina spending as a one-off in projections. It is, however, another reason to be extremely skeptical about the details of many long-term budget projections. (Outside of extremely broad demographic trends such as those facing Social Security without a retirement age increase, and even those can always change.)
As for the Bush Administration bragging about deficits coming down, my problem with that is that it's been happening DESPITE their policies, not BECAUSE of them. Spending has gone up substantially (even aside from war / homeland security) and revenues are much lower than they would have been without the Bush tax cuts.
Discretionary spending reached a low point as a percentage of GDP in 1999 and 2000, at 6.3%. That was almost entirely due to defense cuts-- defense spending was 5.4% of GDP in 1991 when total discretionary spending was 9.0% of GDP, but defense spending had been cut to 3.0% of GDP by 1999 and 2000. The spending cut of 2.6% of GDP terms was thus almost entirely the 2.4% of GDP decline in defense spending.
Since then, we've gone back up to 7.8% overall discretionary spending, with defense being 4.0%. Defense spending is roughly two-thirds of the increase, though certainly not all. The increase in other domestic discretionary spending as a percentage of GDP took place between 2001 and 2002, going from 3.2% to 3.5%. It has remained at 3.5-3.6% each year thereafter. (Unsurprising that it went up when the economy slowed, and then held pace with the economy since, given political incentives.)
Turning to mandatory spending, we see a sharp jump from 2001 to 2002 as well, again because of the slow economy. It went from 10% (roughly the same figure throughout the 90s) to 10.7%, where it has remained in the 10.7%-10.8% band since. The increase has been spread around fairly equally in Medicare, Medicaid, and the "income support" programs of unemployment, welfare, EITC, child tax credit, food stamps, et al. (The EITC may be like a negative income tax, but it is counted as spending by the CBO, along with other tax credits that are refundable and can be paid out in excess of one's tax liability.)
This data comes from the CBO:
The long-term budget effects are clearly dominated by the Medicare drug benefit. Definitely very expensive, though some research suggests that its economic costs are actually lower than predicted, due to particularly strong social benefits.
(Though as a government line-item, it is particularly expensive. It should be noted, however, that there was extremely strong cross-party consensus on some form of drug benefit; indeed, considering the cost-effective nature of pharmaceuticals compared to hospital treatment, excluding drugs from Medicare by itself doesn't make sense, though one may certainly argue about the details.)
Posted by: John Thacker at October 23, 2007 01:25 PM
Nice post, lots of meaty stuff. Steve Conover at Skeptical Optimist pointed out last year that the surplus in the budget at the end of Clinton admin was almost entirely due to cuts in the defense budget.
Posted by: Rich Berger at October 23, 2007 01:35 PM
John T, I am forever amazed at folks who get wrapped around the axle about the deficit. It is simple math. Take all the revenue that comes in. Subtract all the expenditures. The difference is the surplus/deficit +/- number. If the number is negative much of that will have been borrowed from the public, not necessarily all, since some months/weeks run a surplus over expenditures, and borrowing is not needed.
So how do we cut the deficit? Spend less! Spend less than the amount of incoming revenue and we have a surplus, but that is not what has been happening. We have been spending at a RATE less than the RATE of revenue increase. Therefore, we have a deficit that is in retreat, each year since 2004. If this practice is continued, regardless of GDP, that rate will at some point reach zero and might even go positive.
What I see all the time is the mixing and matching of deficit, public debt and Federal debt. In most instances if an economist is doing this they are being disingenuous. Smoke and mirrors. Leftist economists often use the term "General Fund Deficit" or Federal Debt to describe a very large annual number. It is used, mostly, to scare the unknowing into thinking how bad the deficit is. The General Fund Deficit appears to be the total annual liability of the Federal Govt minus annual "tax" (not FICA) revenue.
What is fascinating about the "General Fund Deficit" is it actually includes a surplus. The FICA tax is collecting much more than what is being expended for Social Security, therefore, it is a future liability and counted in that ole "General Fund Deficit." But, the question remains, does that deficit have any meaning?
If my understanding of the "General Fund Deficit" is incorrect please correct me.
Posted by: CoRev at October 23, 2007 02:45 PM
Here's how I see it, and I hope this helps. And I welcome additions, differences of opinion, or perhaps even corrections from others.
Some key considerations when talking about our long-term fiscal health, numbered below just for reference, NOT in any order such as importance.
1) Our current publicly-held national debt-to-GDP ratio.
2) Projected worker-to-retiree ratio over the next few decades.
3) Projected (and unfunded) entitlement expenses over the next few decades under currently structured eligibility, benefits and realistic ranges of assumptions regarding key factors such as medical inflation, along with realistic ranges of assumptions for projected discretionary spending.
4) Annual unified budget deficits, viewed IN CONJUNCTION with the timeline of projected, unfunded entitlement liabilities, and bearing in mind that if deficit reduction is coming primarily from increased FICA revenues and an increased SS surplus (rather than from a narrowing of the General Fund deficit), it only reflects the fact that more future retirees (and entitlement benefit recipients) are paying taxes that are being spent today rather than saved for their retirement benefits.
5) Realistic ranges of GDP growth over the next few decades (obviously very wide ranges, given the levels of uncertainty, but still narrow enough to have optimistic and pessimistic bounds and see the implications vis a vis our unfunded entitlement liabilities and our long-term fiscal health more generally).
Posted by: Brooks at October 23, 2007 07:02 PM
Brooks, this sentence structure makes it tough to decipher:
4) Annual unified budget deficits, viewed IN CONJUNCTION with the timeline of projected, unfunded entitlement liabilities, and bearing in mind that if deficit reduction is coming primarily from increased FICA revenues and an increased SS surplus (rather than from a narrowing of the General Fund deficit), it only reflects the fact that more future retirees (and entitlement benefit recipients) are paying taxes that are being spent today rather than saved for their retirement benefits.There are a couple of things that can be checked.
if deficit reduction is coming primarily from increased FICA revenues and an increased SS surplus ...Key is the ocncept of where the reduction is coming from. Is it FICA surplus, or a reduced spending rate (compared to revenue increase)? Simple math, using the September Treasury FMS Report, shows us a deficit reduction of $85.9B versus an Off budget surplus of $4.9B. Clearly it is a reach to make the "Off Budget" surplus a primary source of the reduction. So it must be from reduced spending rate.
Another misconception is the use of the SS Trust Fund (SSTF) surplus. It was from the originating legislation meant to be passed on to the "General Fund", along with the creation of "non-transferable" bonds to track the surplus amounts. The SSTF is operating today just as it was first intended in 1935. Savings? Of a sort we would allow the Fed Govt to employ. (And that is a complex sentence. Not in structure but in meaning.)
Posted by: CoRev at October 24, 2007 03:42 AM
Sorry, the deficit reduction was supposed to be $85.4B
Posted by: CoRev at October 24, 2007 04:11 AM
Am I reading the chart correctly that the GDP/debt ratio peaked in 1992-93 with publicly-held debt at nearly 50% of GDP? And that even with everything that has happened since, publicly-held debt is still well below the peak--at only 38-39% of GDP?
To me, that seems like good news we should be celebrating. I see the trend lines (based on static scoring I believe) will still not reach that peak for the foreseeable future. That also seems like good news.
I believe the key issue we should be addressing is economic growth. Are any politicians focused on developing policies that are likely to stimulate additional growth?
Posted by: Kurt Brouwer at October 26, 2007 04:55 PM
Kurt Brouwer: You're reading the graph correctly. However, the difference between the Reagan tax cuts of the 1980's and the Bush tax cuts of the 2000's is that we are about to enter into a demographic shift that will increase government spending on transfers enormously, via Medicare and to a lesser extent Social Security. The context matters.
Posted by: Menzie Chinn at October 27, 2007 08:33 AM