Asymmetrical Information

Ditto!

Bartley's WSJ editorial page has pounded states for months on their aggressive riding of the revenue surf from the 1990s and intransigence to accountability. True to form, many states like my home of Ohio are leaning Krugman - more or less chastising the taxpayer and haranguing past leniency, resolving to increase taxes to beat Balanced Budget laws. Most cynically, the legislature (moderate/liberal Rockefeller Republicans, for the record) dredged up a 1936 Use Tax law that had been collecting dust for decades, waving it like the Magna Carta.

I dunno about the wisdom of some of these states. I work in the aviation industry and we provide over $22 million in tax revenue per year to Ohio; for years we've been begging for the state to divert at least $8 million from what is dumped into general funds, in order to sustain infrastructure. Their answer? $2 million. And they still can't keep it in the black.

And we can't repeat often enough the cat Krugman's doppelganger, Brad DeLong, let out of the bag:

more-or-less force federal government spending up from its current twenty percent or so of GDP to somewhere between twenty-five and thirty percent
over the next generation. >>

History suggests that the government never runs a surplus? Who knew?

spending as a % of GDP hasn't topped the 20% mark since 1996, although it's been close and has been decreasing since then to 18.4 in 2000 and 2001. In fact the high water mark for spending since WWII was 23.5% in 1983. I don't know when the Clinton tax increase was passed (92?) but spending was 22.3% in 91 and decreased to the afore mentioned 18.4 in 2000.

I'm not suggesting that the tax cut had anything to do with it. More likely government spending never caught up with the booming economy (and that's a good thing) but it does point out that evaluating spending as a % of GDP is kind of flawed because that means it's ok for the government to spend money when times are good. I would prefer for the government to pay off debt when times are good.

There is no doubt that the government will always spend as much as it can get away with simply because that is how the government wields power. Republican or Democrat, makes no difference when they talk about cutting spending it is always about cutting expenditures to the programs or projects supported by the other guys. Democrats like to spend on programs and Republican like to spend on defense and industry. Direct subsidies to business will come in around 85 billion this year. Or about 4% of revenues. (this is down about 7% 2000 though). Means tested welfare is about 110 billion or about 6%.

My question is what is the incentive for government not to spend? And what is an acceptable level of taxation in a complex technological/industrial economy? Right now the average effective federal income tax rate is about 10.5%. Since the high watermark in 81, 11.1% the lowest its been is 9.8%. Again I don't think tax policy drives this as much as things like the employment situation and other factors. Now if you consider all federal taxes, the burden goes up into the 20% range. But that is driven by corporate profits; more profit, more revenue thus a higher effective rate.

But the discussion has to start with spending, what level of spending is appropriate then after that you have to argue about how to divvy it up among all of us. Maybe per capita is a better way to look at it. I don't have any figures on that.

Are Krugman and Friedman really in disagreement? And if so, over what exactly?

Let’s start with Friedman. He says that he favors tax cuts. Not because of their demand or supply side effects, but because of their impact on spending. Which leads to two questions.

a) Does cutting taxes really cut spending and
b) If it doesn’t would Friedman STILL support those tax cuts?

On point a) I see no evidence. (I’m talking at the federal level. Most states have constitutional constraints prohibiting deficits and so the argument is different with them.) And, curiously enough, Friedman provides none. He simply states the “cutting taxes will lower government spending proposition” without so much as a sentence of corroborating evidence.

A simple look at the CBO tables shows this. In fact one could almost make the opposite argument, that tax cuts RAISE spending. The only period in the last 4 decades that spending fell on a sustained basis was in the 90s after the Bush and Clinton tax increases. After Kennedy’s tax cut spending rose (thanks to the Vietnam war and the Great Society) and Reagan’s tax cut also seems to have had no limiting effect on overall spending. And now, with Bush, we are seeing an increase in spending despite the new tax cuts. And we still have plenty of new spending to go, such as prescription drug benefits.

Oh, and of course, this leaves aside the fact that government mandates and regulations are also not impacted by tax cuts.

So Friedman makes a point favored by many conservatives (“tax cuts will force government to reduce spending”) that simply has no correlation with actual events. And there is good reason for this. Friedman seems to have forgotten that most spending is completely disconnected from the levels of taxation. In 2001 about 67% of spending was mandatory. This includes entitlements and interest payments. Of the $1.8 trillion in spending only about $650 billion was discretionary spending.

On point b) it seems hard to believe that Friedman really thinks any tax cut is OK if it doesn’t result in lower spending. The last time we tried this (under Reagan) deficits were as high as 6% of GDP. Would Friedman be ok with 8% or even 10% deficits?

Finally what is Krugman saying? It’s fair to say he would support more government spending than what Friedman does. But Krugman’s piece is not about the ideal level of spending. His point is that the Bush administration is leading us to ever higher and permanent deficits.

In fact, Mitch Daniels comment pretty much proves that tax cuts do not reduce spending. Even with control of the whole budgeting process Daniels has to accept that we are stuck with deficits for the next 10 years, admitting that tax cut or no tax cut even with the GOP holding both the WH and Congress they can’t reduce spending in any significant way.

GT: disaggregate defense spending. You'll find the answer to your question is "Yes"

Jane,

Even if that is true it is irrelevant to the discussion at hand.

Friedman DID NOT say that cutting taxes would limit non-defense discretionary spending. That's less than 20% of overall government spending.

Friedman said that cutting taxes reduces OVERALL spending not a specific subset of that spending, as you argue.

And the facts don't support his thesis.

Of course if you are now changing the argument to "Tax cuts don't cut overall spending but they do cut a small part of that spending, namely non-defense discretionary spending" I would have to say:

a)So what? It's a pretty small part of the overall spending picture and therefore (relatively) irrelevant.

b) It's probably false as well. Non-defense discretionary spending did fall a bit under Reagan, after the tax cut. But it also continued to fall most of the years after both Bush's and Clinton's tax hike (there was a slight uptick the last two years but not enough to overturn the downward trend). In other words whatever caused the fall of non-defense discretionary spending the tax cuts do not seem to be the cause since it fell even after tax increases.

So your new argument (Tax cuts reduce non-defense discretionary spedning" or NDDS) is:

A) NOT the argument Friedman makes

B) Not particularly relevant since it refers to such a small part of overall spending and

C) Not supported by the facts since NDDS has fallen after tax cuts and ater tax hikes rendering the tax cut explanation a bit useless

GT --

Military spending for the Cold War and the end of the Cold War distort that calculation. It's meaningless to say that Clinton cut spending, when everything else grew while the end of the Cold War allowed him to cut defense spending in half. Overall, if you leave the wars out of it, lower taxes mean lower spending.

Jane,

The one big moral problem with Uncle Miltie's (did you really just call him that?!?) public choice-esque model is that--like most public choice models--it's exceptionally anti-democratic.

First, it seems to assume that government is nothing more than a thief, at best a restrained thief who redistributes with some degree of efficiency. Democracy, in contrast, assumes that the government in some way personifies and tries to promote the collective good of roughly all of the people.

Second, the only normative perscription that we could derive from Friedman's model would almost certainly have to involve lying to the people. Bush can't say: "we're cutting taxes so that we'll have less money and we'll therefore have to spend less to avoid running a big deficit." At least, he can't say this to too many people if he wants to get re-elected.

He has to say something more like "we can afford this tax cut because it will stimulate the economy and provide more revenue to pay for services" or something along these lines. Friedman's model, because it's anti-democratic in its assumptions, just can't be publicly justified in any kind of appealing way; this could only happen if the existing government essentially said "well, we're a bunch of thieves, but at least we're going to be a bunch of thieves who are going to take less of your money."

Congrats on the PBS thing, BTW...

Jane,

If you are going to make an argument you should at least make the effort of sticking to it for more than 2 posts and not meander intellectually all over the place.

The first argument was that tax cuts lead to lower spending. The point Friedman made. No exceptions were made, no specific subsets of spending mentioned.

As I explained above the data does not support this thesis. Do you have any counter data? If so you should post it. If not, intellectual honesty demands you recognize your argument does not hold water.

Instead you came with a NEW argument, that the tax cuts only limited a specific subset of spending, non defense discretionary spending (NDDS for short). I would have expected at least that you would have said "Sorry, that's not the point I was making. I was unclear. Here's the point I'm making " before you introduced a whole new argument.

Never mind that. I looked at this NEW argument (again, let's be clear that that IS NOT the argument Friedman is making or the argument YOU were making in the original post).

And the evidence is unconvincing for that as well. I specifically made three points. Let me reiterate them:

Your new argument (Tax cuts reduce NDDS) is:

A) NOT the argument Friedman makes

B) Not particularly relevant since it refers to such a small part of overall spending and

C) Not supported by the facts since NDDS has fallen after tax cuts and ater tax hikes rendering the tax cut explanation a bit useless

Did you answer ANY of the very specific points I made?

Nope.

You come up with yet ANOTHER argument. One I can't make heads or tails of. Why are you bringing up defense spending? Didn't I SPECIFICALLY exclude that by talking of NON defense discretionary spending?

It's hard to debate with you if you constantly change your argument and don't even recognize it.

So what, EXACTLY, is your point?

Here's the super-exciting OMB data yet again. The general argument, that tax cuts hold down spending, has no supportive data. The specific argument, that tax cuts hold down discretionary non-defense spending, isn't much better: domestic discretionary drops from 4.5% of GDP to 3.9% of GDP in 1982, the first year of Reagan's tax cut. It then slides down to 3.1% over the rest of the 80s, picks up a bit in 90/91 (Bush tax increase), doesn't change when Clinton's tax cut is passed, and then starts sliding again in 1996-00. It then jumps under Bush's tax cut.

So, basically, the 80s and 90s don't support the statement "tax cuts hold down discretionary spending."

The theory I prefer is that "spending levels are democratically determined, and are largely independent of taxation." After all, a deficit is (in general) just an inflation tax; deficit-financed tax cuts should just rearrange the tax burden.

I'm curious what you think of Friedman's theories if you consider him a valid argument-from-authority target, too. And what the heck is that discussion question about? Do economists have an alternative to democracy?

That should be "doesn't change when Clinton's tax increase is passed," obviously.

I think Milton Friedman is a great economist, but he's intellectually being a little dishonest on this one.

Without being as vehement as GT, I would like to point out that over the last 10 years, roughly 2/3rds of Federal outlays have gone to four things: social security, defense, medicare and interest on debt. There is little reason to doubt that these will shrink over the next few years, and lots of reasons to believe that they will grow. Furthermore, if you look at the budget data from the OMB, some of the largest items of what could be termed discretionary spending belong to education and health, both of which are slated to grow given current trends and the increased Federal role in education.

This implies that if you want to cut spending in a meaningful way, you have to cut these items; not having enough tax revenue is not going to be a valid excuse for failing to pay for defense, interest, social security or medicare. So, if you want to cut spending and reduce the size of government, you have to come out and identify things to cut, make coherent arguments for cutting them and try and get teh cuts through Congress. If you want to cut taxes to improve incentives, boost the economy and remove distortions, then you have to come out and make a coherent case for that. The argument that tax cuts are good because will force the government to become smaller is indirect at best, and intellectually dishonest at worst.

A few other points.

1) Jane argues that defense spending under Reagan should be considered very differently from defense spending under Clinton. Fine, but the numbers above don't include the Reagan years, and we are certainly facing a period in the next few years when military requests and requirements are going to be more like the Reagan years than the Clinton years.

2) A country that runs large deficits will eventuially find that the deficits come home to roost, thus the argument that reducing tax revenue will eventually force spending to be cut seems to have some technical merit. The reality, alas, is that when the deficits have become an economic problem, governments respond by raising taxes so as to collect more revenue (see Bush I and Clinton) and narrow deficits. As a result, government spending does not fall that drastically. And even if they do, the point is that only a small portion of spending is likely to fall since the rest is 'non-discretionary'

3) Patrick Sullivan notes that Krugman and De Long have "let the cat out of the bag" by claiming that the size of government will grow to 50% of GDP. he makes it seem as if this is a secret that a) DeLong and Krugman have been hiding this fact b) That DeLong and Krugman WANT the Federal government to grow to 50% and c) That perhaps all liberal economists are like Krugman in DeLong in their wish to see b). I may be reading too much into that last point, but I wish to point out that they are only pointing out the obvious truth: we better start thinking about fiscal policy over horizons longer than 10 years because the projected numbers don't add up. And if my implicit understanding that Mr. Sullivan claimed c) is true, I would like to point out to Mr. Sullivan that it was liberal economists like Alan Auerbach who were poitning this out in the run up to the Presidential campaign when both candidates were blabbing on and on about surpluses knowing full well that the surpluses were tenuous and even if they were not, they would be needed to deal with greater issues down the road.

So I say to conservative economists everywhere: stand up for smaller government, and come up with coherent plans for reducing the size of governemnt, for controlling discretionary spending and tackle the serious issues of exponentially growing entitlement programs. Don't pretend however that cutting taxes, increasing discretionary spending and ignoring entitlment programs will cause a fiscal train wreck that will one day miraculously bring about a world of smaller government.

By that argument, Argentina did the smart thing over the last three years by driving the economy into total oblivion so that substantial reform can now take place from the charred embers. There is a better way to conduct policy than to drive it to a brink of a crisis and then hoep we find the courage to do things better.

Oh, yeah, check out Table 15.1.

State and Local Tax Receipts in 1992: 9.8% of GDP.
State and Local Tax Receipts in 1999: 9.8% of GDP.

State and Local Expenditures in 1992: 10% of GDP.
State and Local Expenditures in 1999: 9.4% of GDP.

More CW-puncturing. State & local spending *fell* over the 1990s, while taxes flatlined. Maybe specific states were "aggressively riding the revenue surf from the 1990s", but they sure as hell weren't doing so collectively.

Next time you're sitting around counting holes in the wall, chart out fed and state spending as year over year % changes of GDP (see bea.gov for numbers). Its not really true that state spending dramatically increased in the 90s.

The states overtook the feds (as a % of GDP) back in the early 70s and their funding levels have remained largely constant ever since (based on my scientific eyeball analysis). What has changed is that fed spending levels dropped like U238 during the Bush/Clinton years.

Politically, Clinton got to champion that he was some sort of financial genius, when in fact, I'd bet this was the work of a Republican controlled congress. Remember all that rhetoric about local control this and local control that?

The problem is when all those fed programs that funded the local guys got cut (in favor of local control), the states didn't have enough to fill the gap -- but they tried. You can actually see a small little uptick during the boom years as states tried to provide services previously provided by the feds.

And guess what? We're in a recession now. Go figure.

(holy smokes, Batman, did I just become a communist?)

Here's a post with links to the state and federal spending, including an analysis by confirmed Friedman acolyte Michael Hodges coming to radically different conclusions than above.

Here's the state & Local data straight from BEA. I wouldn't exactly call it flat.

Here's an extra data wrench in the works. The analysis by Michael Hodges graphs expenditure as a percent of National Income rather than GDP. That's the discrepancy I was looking for. Makes a big difference, doesn't it?

Mindles,

Your commentary on Michael Hodges states that you looked at state spending data going all the way back to '47. Yet my statement on the flatness of state spending clearly indicates that I was measuring from the early 70s (and I got my data from bea.gov too).

There's no question that state spending from the beginning of time has grown significantly, particularly after the war years. But since I thought we were talking about runaway spending since as a result of the dot.com boom, anything going back beyond '71 -- the start of the current state spending trend really isn't that relevant to today's issue is it?

Perhaps I'm missing something, but according to bea.gov state spending was 12% of GDP in 1971 and it was 12% in 2001 -- I don't know if that qualifies as flat in your book, but I can't see why you'd argue with it.

Further as you yourself write, "anyone who's heard the term "unfunded mandate" knows how States have been put in the position of assuming new responsibilities so the Federal government could stay within the range shown."

Aren't we saying the same thing here? That states don't have the money to replace the fed's spending cuts. That spending increased slightly, but not enough to cover the gap suggests to me that the states either haven't the political will or the money (or both) to cover the loss of fed-to-state funding.

I am somewhat confused by Dreck's postings.

1. The graph that shows state government spending as being more than 100% of GDP, what's the deal with that? And it says "straight from the BEA" but links to some random site. I am confused. I hope Dreck can clarify.

2. If you do go to the BEA (or the OMB from Jason's link) you would find that state and local expenditure has increased from 6% to 12.5% of GDP, with virtually all the increase coming before 1975. In other words, over the last 25 years there has been no trend towards state and local expenditure increasing as a percentage of GDP.

His own link shows that Federal government spending/GDP has increased by 6.2 percentage points from 1950 to 1990 but all of it (and even more) comes from entitlment programs. Is this a sign of runaway government or simply a consequence of putting in a government funded pay as you go retirement scheme when there was none before? My understanding is that the phrase "runaway government" typically refers to runsaway discretionary spending, in which case there is none but I could be wrong.

3. The ratio of national income to GDP has remained stable over the last 50 years. So taking the ratio expenditure/national income or expenditure/GDP should make no difference in terms of trend analysis if you use the correct data.

4. The increase in state and local expenditure is driven mostly by Medicare and by education. The creation of a new entitlement program and demographics, coupled with a switch to local funding of education. Once again, not really driven by increased discretionary spending

5. Dreck's graphs do show the killer trends in Federal entitlement programs, and why they need to be addressed. In keeping with the topic of this thread, however, the trends in entitlement programs are not going to be reversed by running more deficits.

Here's a post with links to the state and federal spending, including an analysis by confirmed Friedman acolyte Michael Hodges coming to radically different conclusions than above.

It's an answer to a radically different question. There's been postwar growth in the size of government, yes, but no post-70s growth. I also still see no empirical support for the "deficits starve government" theory, and I have no idea why national income would be a better measuring stick than GDP.

I agree regulations have increased since 1947; but have they increased since 1970? How would we even measure this?

Also, with the bizarre formatting, random bold tags, font color changes, and font size changes from word to word - that's a pretty unconvincing webpage. It's hard to take seriously, regardless of content. It's no Time Cube, mind you, but still.

Final bit: are deficits supposed to have an effect on non-discretionary spending in this theory?

The most striking thing about the comment "In Paul Krugman's model, government spending is given, and tax cuts increase deficits," is that it is TOTALLY unsupported by the quote following, from Krugman, "It's O.K. to run a deficit during a recession, as long as the deficit is clearly temporary. But both the numbers and the administration's search for excuses tell us that there's nothing temporary about the red ink. On the contrary, we'll probably be on a deficit bender until the baby boomers retire — and then it will get much worse."

The numbers that Krugman are looking at are, for the most part, numbers provided by the Bush administration, as it looks (in the person of Mitch Daniels) into the future decade. Krugman seems to project these deficits somewhat further out, although only a few more years, until the baby boomers begin to retire (which could be said to start in as few as ten years).

Where in all of this do you find Krugman asserting that, in general, the cost of government is fixed? If even Mitch Daniels can see only deficits until the end of the decade, surely you have to grant that he, as a member of the Bush administration, is NOT assuming an extravagantly sized government, and is, almost certainly, projecting instead a quite modestly expensive government.

The point Krugman is making is that deficits are tolerable for a short period of time, but are damaging over a very long period. If even Mitch Daniels, presumably a staunch advocate of smaller government, sees no way out of deficits for the remainder of the decade, then it is NOT an issue of too large a government, but of too small a government revenue (taxes).

Krugman is expressing his general disagreement --and perhaps disgust -- with the current administration attitude of dismissiveness toward the possibility of deficits forever.

In this Wrestlemania cage match, style points go to Matt Johnson for 1) staying focused to one or two, yes, points and 2) being n right. The commnets here started bouncing back and forth, stretching the argument like silly putty, so I'm glad somebody called this and tried to direct the discussion toward coherency.

This is needed for a non-expert like myself.

Unfunded mandates are the answer to why states are up the creek. Most states have balanced budget amendments so they are forced to cut, so that would probably explain the flatness of spending over the last 20 years.

You all probably knew that but didn't mention it. Yes increased property taxes mean greater revenue but that only goes so far.

In Washington state at least you have someone, Tim Eyman, who gets the people to pass initiatives that lower fees and taxes and limits the ability for tax increases, with a vote (votes cost money top, ya know). Sounds great, except there are critical areas of the budget which cannot be cut without hurting not just the government people but THE PEOPLE. That is what is happening everywhere.

Tax cuts fuel growth when they are drected toward the right people. How's that for a theory? Rarely does it fuel federal spending. I see no logic there. In fact as the states get healthier I'm sure the federal government feels more comfortable cutting funds off for various areas.

So then when state coffer revenues drop, everything is hit at once. So much for stability.
Your discussion above seemed remarkable lacking of consideration of the human element of theory. That's what journalists do. Now tell where i'm completely !@#$%ed in the head in regard to my thinking here.Still, it was informative.
dimn | Andrew

>>Here's an extra data wrench in the works. The analysis by Michael Hodges graphs expenditure as a percent of National Income rather than GDP. That's the discrepancy I was looking for. Makes a big difference, doesn't it?

The difference between national income and GDP is net income from overseas.

Net income from overseas is unlikely to be part of the tax base for any state, though it might be for the federal government. Gross Domestic Product is likely to match up much better to the states' tax base.

I have no idea why anyone would compare state spending to NI rather than GDP. Or why they would boast about the fact if they had.

dsquared - I didn't use national income in my own post but I disagree with your characterization. When looking at state & local spending in aggregate along with federal expenditures, you could certainly make an argument for using national income. You are correct in pointing out that the Feds would get the bulk of additional revenue from overseas income, but it is an analysis of government expenditures as a percent of the economy, not government revenue.

I'm not sure any of us should be 'boasting'..

I don't know what achilles saw in the graph. When I click on it, it shows state spending heading up to about 12.5% at the max, exceeding clear peaks of close to 12% in 77-78 and 93-94. I made this graph myself from downloaded BEA data. So state spending is flat from peak to peak, but I think it's misleading to say so without looking at the trend over time. Most of the growth is indeed pre-1975, but a 2% of GDP increase from the mid-1980s is nothing to sneeze at.

Could achilles provide me with any statement by Krugman that admits he wants federal spending to increase to 30% (not 50%)of GDP? He's certainly had the opportunity to fess up to it, what with all the NY Times columns he's written about taxes and deficits.

With the exception of DeLong's remarks, the only prominent economist I know of who has said anything like this, is Alice Rivlin in 1993. And her claim was only 25%.

BTW, there have been some very artful uses of numerators and denominators, as well as of the terms "discretionary" and "non-discretionary" in these comments. Taking the latter first, as we business types know, there is really no such thing as "fixed costs", ultimately all costs are variable. The same goes for "non-discretionary spending", or "entitlements". The spending we see today is the result of DECISIONS made in the past. And the baby boomers are soon going to find out just how those kinds of decisions can be reversed (my prediction: raising the age at which SS benefits can be collected along with some kind of means testing).

As for the % of GDP being flat for state and local spending, well that means that, in the late 90s, spending had to have increased rapidly, because GDP did.

I do not believe that DeLong WANTS government purchases as a percentage of GDP to be high, he is saying that given our current commitments it WILL end up being that high unless you change the commitments. There is a huge difference between those two things, and I was quarreling with your statement which seems to imply the former, not the latter. I only brought Krugman into it becaus eyou referred to DeLong being Krugman's "Doppelganger". Given Krugman's background as a top-flight academic economist, I do not believe that he would argue for large increases in government spending as a % of GDP either; once again I would emphasize the difference between wanting something and saying that something will happen unless there is change.

Your point about how we camouflage discretionary and non-discretionary spending is a valid one. Nonetheless, I don't see how you can argue that spending on social security and medicare have to be thought of as different from spending on highways and bridges. My point (and the point of other posters like GT and Jason) was simply that the largest part of the budget is composed of things that are either difficult or impossible to cut by simply starving the government of funds.

Dreck, your graph shows up when I open it as numbers over 100% for state and local expenditure/GDP but it may be a decimal point that got lost in the conversion to a PDF or HTML file. As for the Hodges site, well the less said the better.

Patrick, as for the "artful" use of denominators. Well, i want to point out a couple of things. First that it does not matter whether you take NI or GDP for analyzing trends nothing is going to change given NI/GDP ratio staying constant. Second, that if you want to have serious discussions about private sector crowding out, then you have to do so in the context of ratios to GDP. Third, if you want to argue about levels thats fine, but I doubt that you will find that discretionary spending is the culprit there: if social security, medicare, and interest are the among the most rapidly increasing ratios, you are not going to find a different story in levels. And lastly, if Patrick would like to present all the different components of government expenditure on a real per-capita levels basis instead of ratios to GDP and prove the original claim: namely that the size of government shrinks when you starve it of tax revenue, I would be interested in seeing that.

Would it be accurate to say that reducing revenue streams to the state brings into more stark contrast those activities of the state that society cannot function without, and those that it can? A gigantic fraud was perpetrated when the entire concept of "entitlements" was put forth. "Entitled" to what? The labor and ingenuity of people who have yet been born? Sorry, but no citizen is "entitled" to any other citizen's labor or ingenuity, just as no citizen is "entitled" to the sacrifice of a United States Marine. The state may have a legitimate interest in providing transfer payments, just as it has a legitimate interest in paying the wages of a Marine, but that is much different than saying that any individual citizen is "entitled" to another's labor. Of course, some would cry that it is entirely unfair to means test or cut back benefits, since promises have been hade by previous Congresses, but people who were born in 1984 are not morally bound by the pledge of their labor made in 1965; 18 year olds have a right to self-government, and are not slaves to older people who are, on average, wealthier. If cutting taxes causes people to think about what the remaining taxes are being used for, let's cut taxes even more, and the payroll tax is a fine place to start.

A deeper question is: why should spending be correlated to GDP? Considering that much government spending is some form of social insurance, as GDP rises and unemployment decreases, spending should decrease as a percentage of GDP. Instead, it increased. While the need for defense spending is largely unhooked to GDP, the need for social spending should be inversely correlated to it. So I think GDP is a silly metric to use here.

So taking out that denominator, let's look at something more useful, like population growth. Government spending is largely growing faster than population in all 50 states and in the US as a whole, taking out the Contract with America and a couple of rogue western states. Why should it? Obviously, the aging of America is part of the story, but not the whole part -- looking at the list of major expenditures, such as the giveaway to Denny Rivera and his local in New York, or the sundry new spending programs California enacted when they thought the bubble was going to last forever, confirms my belief as the daughter of a lobbyist that government spends whatever they can.

as has been mentioned, thinking of ss and medicare as "required" spending at a "required, set" level is a good way to end up looking like an idiot, especially with a big demographic problem.

if one wanted to scale back these programs (say for efficiency reasons, improving price sensitivty amongst medical consumers and retirment investment..) one may decide to bring the point of pain forward by several years (since these programs will crash in a veryshort time, only question is when)

when the pain is evident and visible, people will be more amenable to drastic measures (so we can raise retiement age to 120, cut benefits by 75%, privatise it....) since there ain't any money in the till

there is no constitutional requirement to spend money on anything except defense and courts, and even there there is no explicit funding formula

please keep this in mind when one starts to think that there is "nothing" that can be effectively cut from gov't spending...

This seems a bit like the debate over whether deficits determine interest rates - lots of factors to obscure the point. In the case of deficits and interest rates, growth and inflation make a big enough difference to overshadown the impact of deficits. In the "starve the beast" argument, trouble comes because:

- entitlement spending is "different"

- defense spending is "different"

- demographics are powerful

- offcial spending rules are powerful (says Greenspan)

- surprising revenue growth can make a declared preference for fiscal rectitude pretty painless (under Clinton)

- especially when the declared preference for fiscal rectitude was not-so-secretly helped along by a plan for slower defense spending initiated under another Bush by Dick Cheney

Amid those factors (surely I have missed some), there is room to hope that government can be starved into submission, but finding incontrovertable evidence is, apparently, not so easy. Facing up to demographics and making explicit policy choices seems a more certain way to avoid fiscal troubles down the road.

A deeper question is: why should spending be correlated to GDP?

Maybe. Maybe not.

You posted an article by Friedman and very clearly said you agreed with his point of view. But Friedman does not discuss whether measuring spending as % of GDP is correct or not.

When comparing government spending across time the commonly accepted practice is to calculate it as % of GDP. Using these numbers Friedman's point, that tax cuts reduce spending, does not seem to hold true.

Now maybe you have developed an alternative metric. If so, I'd be curious to understand exactly what it is and what it tells us about the question at hand.

But he doesn't discuss GDP at all, GT. He just says that if revenue increases, so will spending. I agree. And I think the data does too.

Jane,

It seems you've come up with yet ANOTHER argument.

Here's what Friedman says:

how can we ever cut government down to size? I believe there is one and only one way: the way parents control spendthrift children, cutting their allowance. For government, that means cutting taxes. Resulting deficits will be an effective--I would go so far as to say, the only effective--restraint on the spending propensities of the executive branch and the legislature. The public reaction will make that restraint effective.

Friedman says cut taxes and that way you will limit spending.

Now practically every economist I have ever read on this topic has always looked at % of GDP when trying to measure this.

That's why I used those numbers and came to the conclusion that Friedman (and you) were wrong, since tax cuts do not seem to have limited government spending at all.

To this you responded by changing the argument, saying that what it limited was not spending but a specific subset of that spending. Shall we call it Jane's Point version 2.0?

Now comes Jane's Point Version 3.0, which, as best I can understand it says that using % of GDP is not the correct way to measure all of this and some alternative measure (which you haven't defined) would prove you right. Although even there you are not clear as to what would be proved right, your very first point or Version 2.0.

I believe you mentioned in your PBS interview that the problem with many journalists is that they lack basic training in statistics.

Give that, I ask again? What, EXACTLY, is your point? How, EXACTLY do you think it should be measured and why?

GT, it's the same point. Revenue up spending up, except for non-discretionary items like defense. I'm sorry if you can't understand it, but I can't make it any clearer.

Achilles - sigh - yup. I see it. In a different browser I can't see the decimal at all. Well, as they say, "assume a decimal..."

Re. the appearance of Hodges pages - I know what you mean, and I do find it hard to read. However, as an equally design-challenged individual I try not to pre-judge the author by his HTML. I remember having a similar reaction to U.S.S. Clueless when I first saw it in the summer of 2001 (it had a different format then). I'm very glad I took time to look harder.

Hodges takes his analysis from the same place as the rest of us. While his style is certainly odd, I consider him a basically an issue-blogger.

GT, your attempts to provoke without illumination are quite tiresome. Jane can respond to your posts without my aid, but out of being bored with the the tone of your remarks, I will presume to butt in.

If we assume that the ability by the state to borrow is not infinite, then how much the state can spend is limited by X, tax revenues, and Y, issuing bonds. Thus, X plus Y= total spending. If X is reduced, the total ability of the state to spend is reduced, unless one wishes to make the case that reducing taxes increases the total amount of borrowing the state may engage in, which would be an interesting argument, to say the least. The figures you present cannot address the important question, which is what spending would have been if tax rates had not been reduced. Of course, one cannot prove that failure to reduce tax rates would have increased the total amount spent, either. As a matter of record, however, how often have non-defense spending programs been cancelled, or significantly reduced? Does not the very existence of programs which automatically increase every year, absent actual ongoing legislative action, prove the inclination of spending to increase to funds available, whether through taxing or borrowing? Can you cite an example of a program which has a similar bias towards decreased spending every year?

"Revenue up spending up"?

That's your point?

Oh well.

No definition of what spending is or revenue is, how you measure across time, how you adjust for real growth and inflation.

How do you define "revenue up" and "spending up"?

Are we on version 4.0 by now?

Will,

Maybe you can help Jane.

What EXACTLY is the point being made.

Friedman is claer about it.

He says cutting taxes is important not because of demand or supply effects but BECAUSE IT WILL LIMIT SPENDING.

He doesn't define what he means by that and he provides no proof or supporting evidence.

I looked at government spending and revenues as % of GDP and I find no such link.

Now maybe you think we should use another statistic, and not % of GDP. If so, which one?

Or maybe you think that tax cuts limit some PART of government spening but not al of it. If so, which part?

By the way Jane, defense is not a "non-discretionary" itme. It is, in fact, part of discretionary spending.

Jane asks "Why should spending be correlated to GDP?". I can only speak authoritatively for myself (but I would guess that GT, Patrick, KHarris et al. would agree with me) in saying that the point is not that spending SHOULD be correlated to GDP but that it IS correlated to GDP over the last 25 years.

Jane then goes on to say that since "Government spending is largely growing faster than population in all 50 states and in the US as a whole" this is evidence of runaway spending, again judging from from her anecdotes runaway spending of the non social insurance, non-defense, non-interest kind.

Instead of assertions based on anecdotes, perhaps Jane could in fact back up her claim and show that increases in real per-capita government spending come from the Denny Rivera giveaways instead of from payments to Jack and Jill Retiree, payments to bondholders and defence spending. I agree with her that controlling the level of discretionary spending instead of discretionary spending/GDP can help limit the size of government but in the long run this is only going to have a minimal impact. You can freeze all discretionary payments at their current levels, but unless you change the big 4, nothing is going to happen over the next 50 years other than government getting larger.

I have never seen any introductory economics book that claims to have a model of government spending. Friedman posits such a model, government is contrained by the size of its revenues, but alas I don't think the model is right.

I am encouraged by today's news reports which says that the President will talk about Medicare and Social Security reform in the State of the Union address.

>> I also still see no empirical support for the "deficits starve government" theory

The Reagan deficits resulted in Congress adopting "paygo" spending rules that forced all new discretionary spending to be financed -- and since Congress enjoys paying much less than spending, that cut its growth to just 1% a year through Bush I years and well into the Clinton years.

But *immediately* upon the appearance of a surplus in 1998, the rules went out the window and discretionary spending exploded. In fact, DeLong has a conversation about this with former CBO head Rudy Penner on his web site....

"So what happened at the beginning of 1998 to change things so completely? That's still not clear to me...
"Rudy: I believe it was the surplus. PAYGO was originally designed to stop tax and entitlement policy from increasing the deficit. (Really, to preserve the gains from the 1990 budget agreement.) After 1997, it had the effect of preventing any reduction of the surplus and that didn't make much sense...."

Rudy well knew what "makes sense" to Congress. ;-)

From then through 2002 discretionary spending rocketed up by more than 8% a year in a happily bi-partisan manner. [And for the record, as Clinton signed off on it and took his share of that 8% for the first two years, more than all of which was attributable at first to the SS surplus, I don't remember Krugman ever saying "peep" about that squandering of same. That seems to have become a much more important to him the day after election day 2000. ;-) ]

I agree with Will and LUA about the dangers of thinking about social security and medicare as entitlement programs. But realistically, what are the odds that they are anything other than entitlement programs today? No administration, Republican or Democrat, will dare cut off current retirees from their social security benefits. The best bet then is to break the link for the next generation of retirees, which is what the Administration wants to do, but the problem is that then you have to come up with the funds to pay the current generation of retirees (which the Administration has ignored). This is what Paul Krugman harped on for a long time last year. Political courage is required, and action needs to be taken now. All the talk about starving out social security by reducing government revenue is somewhere in the spectrum between passing the buck and wishful thinking.

A couple of other responses:

Mindles: I see now exactly where the decimal point should go so I am sorry for making it seem like you had gone off the deep end. The numbers above 100% threw me for a loop.

At the risk of sounding patronizing, I have to say that I like the intent of Hodges and I think his basic message that this generation may be screwing over future generations is correct. But I would not rely on it for authoritative analysis. I also think he tries too hard to reconcile this belief of the world with his political biases, but thats my personal opinion.

Will has a nice little model where X = tax revenue and Y = borrowing.

"X plus Y= total spending. If X is reduced, the total ability of the state to spend is reduced, unless one wishes to make the case that reducing taxes increases the total amount of borrowing the state may engage in, which would be an interesting argument, to say the least."

But in his zeal to attack GT (which I assume has a historical context that I am not aware of given that I am a recent arrival here) he ignores that several posts (including by GT) have shown that the link between X going down and the right hand side going down seems unsupported by data and is likely to be even less supported in the future. Its at the STATE level where the ability to issue Y is constrained that the link between X and the right hand side variable becomes stronger.

Well, actually Milt proposes a model "expenditures = revenues + whatever the government can borrow," where "whatever the government can borrow" has a big fudge-factor associated with it.

Over the short term, expenditures aren't limited to revenues, but over the long term you reach a point where you can't borrow any more and/or can't service your existing debts (therefore nobody will buy more debt). (In other words, the long-term model is expenditures = revenues + new borrowing - debt service, where debt service is a function of past borrowing.) So the "whatever they can get away" with part diminishes in time, but not until you've run up an N% of GDP national debt (where N is a number that an economist could identify better than me). At that point, your country is screwed and ends up like your favorite economic basketcase of the 20th century.

A deeper question is: why should spending be correlated to GDP?

Education, infrastructure, research, agriculture, and justice should all track with GDP.

Defense spending may be subject to one-time shifts due to the rise and fall of various threats, but over the long run it should track technology growth (for weapons outlays) and income growth (for salary outlays).

the need for social spending should be inversely correlated to it.

You seem to be using the "social spending = AFDC welfare" shorthand here, which is not even remotely correct. Poverty-specific outlays are a tiny fraction of the budget. "Social spending" in the US budget, if it means anything at all, means the virtually-everyone-gets-them programs like SS and Medicare, which should increase with income.

Most of the growth is indeed pre-1975, but a 2% of GDP increase from the mid-1980s is nothing to sneeze at.

Over 20 years, while the size of the federal government went in the other direction. Neither of the statements "government growth is out of control" or "government tends to grow unchecked" are accurate; total government size in the United States has been pretty constant since the 1970s.

Would it be accurate to say that reducing revenue streams to the state brings into more stark contrast those activities of the state that society cannot function without, and those that it can?

The "deficits cut spending" argument is fundamentally anti-democratic. Friedman, et al. apparently think, for whatever reason, a democracy is simply incapable of cutting spending unless its tricked into doing so by high deficits.

The premises aren't true: government spending doesn't "tend to grow uncontrollably" (it's dropped a bit as share of GDP), and specifically it hasn't done so since the 1960s. The evidence doesn't support the argument, either: there's no visible link for tax cuts->deficits->cuts in spending levels. I also think the logic (deficits should result in spending cuts, not tax increases) is pretty bizarre, too; why are you so certain tax increases, or just ignoring the deficit, won't be the results? What kind of argument is this?

If you have to desperately throw in controversial patch after patch to your model that you didn't think were necessary at the start (national income, spending shouldn't track GDP, selective uses of inflation), maybe you should re-examine your model.

GT, since he doesn't define what he means, trying to prove him wrong is pointless, and your figures are pointless also, since they do not answer the unknowable, which is what spending would have been if taxes had not been cut, or what GDP would have been if taxes had not been cut. Counterfactual history can be entertaining, but is of limited utility. What I mean is that if government spending is either accomplished through tax revenues or borrowing, and we assume the state's ability to borrow is not infinite, then reducing tax revenues reduces the ability of the state to spend, unless it can be shown that reducing tax revenues increases borrowing capacity. This is nothing more than common logic. If one concedes that there is a built-in bias towards increased spending on the non-defense side (due to what is poorly labeled "non-discretionary" spending), and defense spending will occillate, depending on the circumstances, then reducing tax revenues forces the state to eventually increase borrowing (keeping out assumptions of economic growth, which also reduces this model's utility; like most things in a modern economy it really becomes too complex to accurately reduce to simple assertions). If we assume the ability to borrow is not infinite, then reducing tax revenues inhibits the bias towards increased spending. If one believes, as I do, that many of the activities of the state are morally illegitimate uses of force, then one favors reducing taxes, since eventually the ability of the state to borrow is constrained, and at last a discussion may be had as to what really is required of the state.

Comments on various comments...

>> The difference between national income and GDP is net income from overseas.

Oh, not so. That's the little difference between GDP and GNP, which is slight. But GDP goes through a bunch of adjustments to make NI, which is about $2 trillion less. That ain't chump change. See Table 1.9 in the Frequently Requested Tables at http://www.bea.gov/bea/dn/nipaweb/index.asp for details.

>> Net income from overseas is unlikely to be part of the tax base for any state

Huh? NY isn't going to tax me on what the Feds do? So I can invest abroad as a free and easy state tax shelter? Why is NY going to be so much more generous than the Feds?

>> My point (and the point of other posters like GT and Jason) was simply that the largest part of the budget is composed of things that are either difficult or impossible to cut by simply starving the government of funds.

I dunno. On the one hand it seems very "hard" to reform Social Security. OTOH, SS's financing problem could be solved in a day just by not paying benefits to Warren Buffet, Bill Gates' father, and the various other "rich" who collect it. Any by reducing benefits paid to the many, many more "far from poor" who get it. They wouldn't be hurt -- in fact, Gates' father is running around the country saying he should pay more in taxes -- so it wouldn't be hard economically *or* socially to fix SS.

It's just that *politically* many people are set in concrete against means testing, in order to preserve the whole system "as is", even at the cost of paying current transfers from low-income workers to millionaires.

But even the politics here could change *easily* in the future. If Krugman gets his wish and SS starts getting large transfusions from general revenue, it'll become easy for people to say: "Millionaires are getting retirement subsidies not from their own SS contributions but from *general income tax revenue* -- while there's not enough to pay for welfare for the truly poor, or for schools, or for health care for me ... and we are running huge deficits!!" What comes next?

In fact, it's much better of course for millionaires to get subsidies from general tax tax revenue collected through progressive income taxes than from regressive payroll tax revenue.

So suppose people started generally realizing and saying right now: "Millionaires are getting subsidies from regressive payroll taxes imposed on the working poor!" What then?

A lot of issues that are easy economically but difficult politically aren't even that difficult politically, necessarily -- a whole lot depends on who succeeds in framing them in what way.

>> A deeper question is: why should spending be correlated to GDP? Considering that much government spending is some form of social insurance, as GDP rises and unemployment decreases, spending should decrease as a percentage of GDP.

Well, Social Security was created to help the poor elderly out of poverty in world were there were no reliable private retirement plans.

So in a much later world where people are getting richer and richer by the year, with the elderly are the richest of all, and with assets in private retirement accounts accumulating by the trillions of dollars, spending on Social Security ... goes up of course! And the same for Medicare, in fact *much more* so.

Public choice economics explains it clear as a bell -- and makes nice predictions for the future too, for anyone who cares to look.

Jason, why is it more anti-democratic to obtain a vote by promising to allow the voter to keep more of what they have obtained via mutual agreement, while failing to address what borrowing might thereby be required, as opposed to obtaining a vote by promising to forcibly take property from another citizen, and give it to the voter, while consistently underestimating the amount of taking that will be required to fufill the promise? Or is it your position that those who have advocated new spending programs since the New Deal have accurately predicted what the costs would be, or, even more implausibly, it is mere coincidence that the estimates nearly always come in low, as opposed to high?

Jim Glass Writes

"But *immediately* upon the appearance of a surplus in 1998, the rules went out the window and discretionary spending exploded.....From then through 2002 discretionary spending rocketed up by more than 8% a year in a happily bi-partisan manner."

Nice story. It would be nicer if it was true.

% increase in NOMINAL Discretionary spending [1992-2001] from CBO

Discretionary Non Defense Domestic
1992 -5.50% 0.09%
1993 -3.43% 1.04%
1994 -3.52% 0.37%
1995 -3.13% 0.64%
1996 -2.82% -2.26%
1997 2.12% 2.69%
1998 -0.55% 0.89%
1999 1.94% 3.54%
2000 6.84% 7.22%
2001 3.69% 5.46%

so unless by 1998 you meant 2000 I have to say the 8% skyrocketing seems to not be evident.

Jim Glass also writes

"And for the record, as Clinton signed off on it and took his share of that 8% for the first two years, more than all of which was attributable at first to the SS surplus, I don't remember Krugman ever saying "peep" about that squandering of same. That seems to have become a much more important to him the day after election day 2000. ;-)"

Nice story but again the facts are not so nice.

Surpluses as % of GDP [1992-2001]

On-Budget Surplus Social Security surplus
1992 -5.5 0.8
1993 -4.6 0.7
1994 -3.7 0.8
1995 -3.1 0.8
1996 -2.3 0.9
1997 -1.3 1
1998 -0.3 1.1
1999 * 1.4
2000 0.9 1.6
2001 -0.3 1.6

So unless by Clinton you meant Bush, there was no squandering there. The falling deficits were ON-BUDGET, not because of booming social security receipts.

It is amazing to me how much people will be willing to go to discredit Clinton. Now there are valid arguments about stock bubbles, and capital gains taxes that can be made regarding the source of those surpluses, but the facts as recited by Jim don't match. Perhaps my quick reading of CBO data is wrong, i look forward to finding out my mistake.

Will,

It's not pointless (or at least no more so than any other debate we have here). Yes, we have incomplete information. So? That's always the case. That's why we make some assumptions and try to clarify them.

And if it is true that we can't know exactly what Friedman has in mind I CAN ask Jane what SHE has in mind when she says she agrees with Uncle Miltie.

You propose a different variant than what Jane says. You say that government spending is either accomplished through tax revenues or borrowing, and we assume the state's ability to borrow is not infinite, then reducing tax revenues reduces the ability of the state to spend, unless it can be shown that reducing tax revenues increases borrowing capacity.

Three problems or counter arguments.

1) As you point out it may be the case that reducing tax revenues increases borrowing capacity. This could happen in several different ways including people having more money they can then lend to the government. The only way to show if this is true and to what extent would be to look at the empirical data. Do you have any information? I don't.

2) Even if you discount point 1) your analysis depends on the ACTUAL level of spending being close to the THEORETICAL limit, as determined by the sum of taxes and borrowing. Let's call that the maximum level of spending (MLS). But if the ACTUAL level is far below the MLS then you could cut taxes (and reduce MLS) and STILL raise ACTUAL spending because the available spread was so big. Suppose that the MLS (as defined by you) was 30% of GDP. And ACTUAL is 20% of GDP (using round numbers to simplify). Then you cut taxes by 2% of GDP so that MLS is now 28% of GDP. You can still raise ACTUAL spending significantly.

What does the data show? Well, Japan has run deficits as high as 10-12% of GDP. The US ran very high deficits back in WW2. We can easily top the 6% that we reached under Reagan. So even if we agreed with your model there still seems to be plenty of room to raise spending.

Oh, and I’m not even addressing the fact that your definition of maximum spending depends on the government not simply printing money and creating inflation. This is CRUCIAL. In that case MLS would be even higher.

3) Finally there is the little issue of the data. The fact remains that tax cuts have NOT resulted in spending falling, as your model would predict. I am suing the % of GDP numbers for this. Do you have an alternative statistic, like I asked Jane? Maybe you do, and using that number you CA show that tax cuts reduce spending. If so, care to share with us what that definition is and what the numbers say?

I apologize to achilles for neglecting to include the url for Brad DeLong's comments:

http://www.j-bradford-delong.net/movable_type/archives/001391.html

Had I done so (as I had in an earlier post on another thread) he would have seen (and probably not have wasted his time arguing otherwise) that DeLong is definitely expressing HIS PREFERENCE. The sentence of his that immediately followed what I did quote being:

" [Steve]Cecchetti's belief that taxes should be restricted to 19 percent of GDP and debt to 50 percent is a decision to award victory to one side of American politics in a debate that has yet to be started."

Achillies, perhaps I am misinterpreting you, but is it your assertion that national government has an infinite ability to borrow; that it could endlessly finance it's activities soley through borrowing? If not, then it logically follows that reducing taxes reduces the ability of government to spend money.

As an occasional reader of De Long's page, I did see the Ceccheti piece. Delong's full quote is

"I think his particular framework is a bad idea: the existence of the baby-boom generation, rising medical costs, and the belief that everyone ought to be able to see a doctor together more-or-less force federal government spending up from its current twenty percent or so of GDP to somewhere between twenty-five and thirty percent over the next generation. Cecchetti's belief that taxes should be restricted to 19 percent of GDP and debt to 50 percent is a decision to award victory to one side of American politics in a debate that has yet to be started."

Note the following:

1. DeLong says that the reality (not his choice but the reality) given demographics and medical costs is that G/GDP will be between 25% and 30% in the future.

2. Ceccheti's argument is that tax revenue be restricted to 19% of GDP (not spending but tax revenue) and debt to 50%.

3. DeLong does not think Ceccheti's idea is a good one. But Ceccheti's idea is not about the ideal size of government, it is about the ideal amount of tax revenue. DeLong thinks that restricting taxation in this manner without targeting spending is a stupid idea. And I agree. Don't you?

In fact he closes by saying

"But the broader point that fiscal policy has been a total botch and that major plans for reform are desperately needed is completely correct. Ideas, anyone?"

So I stand by my statement that at no point does DeLong argue that government spending SHOULD be 25-30% of GDP. He argues instead that politicians should pull their heads out of their ass and start tackling problems instead of coming up with cockamamy theories about starving those problems away. I agree wholeheartedly

Will, no i don't think that the Federal government has an infinite ability to borrow. in fact I would aver that the amount that I think the Federal governments ability to borrow without causing harm is less than what Glenn Hubbard thinks! My point is that when faced with deficits they have continued to borrow regardless of whether it is a smart idea or not. Their willingness to keep borrowing is what causes your model to break down.

At the risk of sounding like even more of a broken record than I currently do, what I am saying is that

1) When tax revenues fall, government typically borrows instead of cutting spending. [Reagan era]

2) When they feel like they have borrowed too much or borrowed for too long they raise taxes instead of cutting spending [Bush I and Clinton].

Given this history, I think Milton (and you and Jane) are wrong in clinging to a model that does not seem to hold in either the face of past experience or future reality.

Will,

It seems achilles and I are saying the same thing about your model but from different perspectives.

Achilles points out how the data does not follow what your model would predict. As he says politicians have preferred to borrow more and, eventually, raise taxes, rather than cut spending.

I tried to explain that the reason your model fails is that it assumes that the actual level of spending is very close to the theoretical maximum. That's the only way that your model works.

But that doesn't seem to be the case. There's plenty of 'room' for the government to borrow more money so they can easily cut taxes and raise spending.

GT, I could assume that you are asserting that you are Napoleon, and base an entire post on that assertion, but I don't think it would be particularly enlightening. Friedman's response in an interview is not exactly a dissertation, so I'm just saying that our assumptions as to what Friedman meant can be entertaining, I guess, but aren't exactly enlightening. As to your third point, again, you fail to understand that you are engaging in counterfactual history (Friedman may be also) in that you are assuming what spending would have been IF taxes had not been reduced. You have, and none of us have, any way of knowing what spending would have been without the reduction in taxes, just as we have no way of knowing what domestic spending would have been if defense spending had not been increased. All your statistics show is that spending was at a certain level following tax reductions; this does nothing to tell us what spending would have been absent the tax reductions. I did not predict total spending, or any type of spending, would decrease with tax reductions. I simply said that reducing the amount of tax revenue acts as a constraint on spending. The two statements are not synonymous. Yes, national government can borrow, although not infinitely, and, as you rightly note, it can print money, although not infinitely. If we assume that borrowing and printing money have finite capacity, then reducing tax revenues reduces spending capacity. True, actual spending may run short of capacity, but I still think it useful to reduce total capacity, since I view much of what the modern state engages in to be a morally illegitimate use of force, while plainly conceding that most people disagree.

As to the politics of the matter, yes, I would prefer that spending be reduced substantially, starting with means testing for Social Security and Medicare. I see no difference, however, between obtaining votes by promising tax cuts while failing to address what the implications of borrowing or printing money will likely eventually be, and obtaining votes by promising transfer payments, while failing to address what the true cost of those payments will likely eventually be, and since I consider much of the transfer payments to be morally illegitimate to begin with, I find the latter activity more objectionable.

GT is back to playing with cash basis numbers, I see. Let's take another look at the CBO numbers he provided for us in a different thread. All these are % of GDP:

Year Revenue Spending Total Debt

1974 18.3 18.7 23.8
1975 17.9 21.3 25.3
1976 17.2 21.4 27.5
- - - - - - - - - - - - - - - - - - -
1977 18.0 20.7 27.8
1978 18.0 20.7 27.4
1979 18.5 20.1 25.6
1980 18.9 21.6 26.1
1981 19.6 22.2 25.8
1982 19.1 23.1 28.6
- - - - - - - - - - - - - - - - - -
1983 17.4 23.5 33.0
1984 17.3 22.1 34.0
1985 17.7 22.9 36.4
1986 17.5 22.5 39.6
1987 18.4 21.6 40.6
1988 18.1 21.2 40.9
- - - - - - - - - - - - - - - - - - -
1989 18.3 21.2 40.5
1990 18.0 21.8 42.0
1991 17.8 22.3 45.4

About which numbers I note, 1. The revenue numbers for the first three years 74-76 are entirely typical of the period from 1962 through 1976, in which revenues were below 18% of GDP 11 out of the 15 years. While the spending for 74-76 is by a Democrat controlled congress (and would have been higher without the veto pen of Gerald Ford).

2. note the revenue in the six years from 77-82, because it is aberration. No year is under 18%, and two years are over 19%. There was no similar period in any of the years 1962-93.

And then note the trend in spending; increases up to a peak in 1983 at 23.5%.

3. In the next six year period, we returned to a more normal (1962-76) level for revenues, and, in fact, the upward trend for spending was stopped, and then reversed. Until Reagan left office.

Which picture is why Tip O'Neill bitterly complained that "Reagan's deficits" hindered his ability to spend money.

Will,

Not sure what that Napoleon comment was all about. If you are intellectually unable to engage in arguments that require making assumptions you limit severely the scope of what you can talk about.

And you fail to realize that I have not been asking Friedman what he thinks. I have been asking Jane who has had ample chances to detail what she means by "revenues up spending up".

As for your point it is, as you now make clear, not related to what I had been arguing.

As you post: I did not predict total spending, or any type of spending, would decrease with tax reductions.

I realize YOU did not predict that. But both Friedman and Jane DID predict that. Admittedly we don't know exactly what Friedman had in mind (which in effect makes Friedman’s article useless) but we can guess. As for Jane, I have asked repeatedly to explain what SHE has in mind.

So then, if you are not arguing what Jane or Friedman are arguing what is your point? You say:

I simply said that reducing the amount of tax revenue acts as a constraint on spending.

How so? We've been through this and both Achilles and I made specific counterarguments which you have not addressed.

Let's repeat them:

1) From a THEORETICAL point of view your argument is based on the idea that total spending by the government cannot be more than Taxes + Borrowing. To begin with this is false since the government also has the option to print money and, history in many other countries shows, could do that for many years.

But let's leave that aside. Let's assume you are correct and Maximum Spending cannot be more than Taxes + Borrowing. So, does that mean that cutting taxes reduces or limits spending? This is the core of your argument.

It turns out the answer is NO. Why not? Because you confuse two separate concepts. One is Maximum Spending and the other is Actual Spending.

All your argument says is that Maximum Spending may be limited. But what everybody here has been talking about is ACTUAL spending. What the govt has actually spent not the maximum it could theoretically spend. This is a crucial distinction. The fact that Maximum Spending may be limited by tax cuts tells us nothing about what that means in real life. The ONLY way that would have any practical consequences is if our current spending is close to this theoretical maximum.

We can easily borrow much more than the tax cuts as we proved in the 1980s and we are doing today.

What then is your point? Where you making simply a comment about theoretical limits without any relevance to what is happening today?

2) From a PRACTICAL point of view the numbers remain as solid as ever. Looking at % of GDP it’s clear that tax cuts have not limited spending and tax increases have not necessarily raised them. Maybe you have some alternative statistic which you consider more relevant. If so, which is it?

Finally, your argument that we can’t know with certainty what would have happened in the absence of the tax cuts while true is useless. We can never know with certainty anything for that matter.

If Friedman or Jane or anybody else says that a reason to cut taxes is that it will limit spending I think it is fair to ask what the proof of that is. Looking at % of GDP figures we notice that tax cuts have not reduced spending and tax increases have not always increased it. The data seems to show that there is no strong relationship between cutting or raising taxes and future levels of spending.

Will, Jane or anybody else,

Since long posts tend to confuse some let's make it simpler.

How about this:

1) What, EXACTLY is your point? What, EXACTLY, are you saying about the relationship between taxes and spending? Try to 'operationalize' the variables, meaning try to make sure they are defined in a way that can be easily measured.

2) What data are you basing that on? What do the numbers say about your thesis?

And to make it simpler let me offer my point.

I think that the idea that cutting taxes will limit or reduce government spending is false.

I am talking only about the federal government. I am looking at overall spending, not a subset of it or even the internal composition of it.

I base it on the CBO time series using % of GDP numbers for both revenues and expenditures. I use that time series because that is what is normally used for these time comparisons.

Therefore I think that whatever the merits of Bush's tax cut (or any other for that matter) limiting or reducing overall spending is not one of them. The historical evidnce, using the variables I indicated, is clear that whatever drives spending decisions tax cuts or tax increases do not seem to be the reason. There has been increases in spending after tax cuts and decreases in spening adter tax increases.

GT, I have already stated that that you were correct in saying that printing money was an option available to the state to allow it to spend. However, printing money, like borrowing, is not an infinite option. If three variables are available as a means to allow spending, and one variable is reduced, and unless it is asserted that that reducing the one variable increases the total potential of the other two, then it can be said that reducing the one variable acts as a constraint on spending. For the final time, this is NOT synonymous with saying that reducing that variable means that spending will decrease. It is not synonymous, because constraints are often overcome, and it cannot be predicted what elected officials will do in regards to the other two variables. If one wishes to reduce the ability (total potential) of the state to spend, however, eliminating or reducing whatever means are available to the state to do so has that effect. If a Constitutional Amendment prohibiting borrowing (I'm obviously not advocating this) were passed in conjunction with tax cuts, total potential would be reduced further, since then the only option available would be to print money. If you have three options available by which to accomplish actvity A, the act of reducing or eliminating those options acts as a constraint on your ability to engage in that activity. Whether that constraint will be overcome is dependent on your willingness or ability to engage in the remaining options. Now, as we approach the politics of the matter, any examination of campaign rhetoric shows that the advocacy of large amounts of borrowing carries with it a political cost, unless it is assumed that candidates are completely irrational, and bloviate on the subject for no particular reason. Likewise, increasing taxes carries with it political cost, as countless politicians have discovered. Of course, advocating spending cuts carries gigantic costs also. Eventually, something has to give. I tend to favor actions which bring these conflicting positions into focus, so to the degree that decreasing taxes does so, and because much of what the taxes are funnelled to I consider to be morally illegitmate instances of force, I favor almost any tax decrease. Yes, I want a crisis, the sooner the better. The Bush proposal is too mild for my tastes; I would prefer a 33% reduction in payroll taxes for a start.

The a priori plausibility of the theory attributed here to Friedman makes it, I think, worth going to a certain amount of effort to try and patch. (The alternative hypothesis that government does NOT moderate its spending in response to deficits is, in a way, more damning of the democratic process than any idea Friedman has ever come up with.) Whether the patching efforts have been successful is doubtful; the fact that relatively little government money nowadays goes to one-shot expenditures rather than ongoing programs makes the theory difficult to test.

OK Will, so you are focusing on the theoretical argument.

As I explained your model, although technically TRUE, is useless in the practical world because even if it constrains spending it does so at levels much higher than current spending (since we have so much more room to borrow). If before a tax cut we could only 'pay' for 30% of GDP spending, and a tax cut reduces that to 28%, that is still much higher than what we spend today.

If we apply your model to the current levels of taxes and spending it shows that we can cut taxes and still have no impact on spending. Which was my point from the beginning.

Unless, of course, you claim that we are already at the theoretical maximum.

btw, in the real world things don't work like in your model. Why? Because there is so much room to borrow.

Your model may apply to state governments. But I'm not sure.

It just doesn't apply to the federal government because of its huge capacity to borrow. So when taxes are cut they can still raise spending and cover the difference with more borrowing. Over time this leads to higher taxes. That's what happened under Reagan. And what seems to be happening under Bush today.

Achilles amuses...

1) In his rush to refute the numbers I posted showing the increase in discretionary spending that occurred promptly upon the arrival of the surplus (which came from analysis by Morgan Stanley, and I shan't expand on here) he gave us his own numbers for average spending growth, which work out to :

Pre surplus: -2.4% 0.49%
Post surplus : +4.16 5.50%

The latter being an 11-fold increase (I won't multiply the negative number). Very good! Enough for me. From now on I won't use MS's numbers, I'll use his.

2) Somehow, while he clearly thinks he knows more about what happened upon the arrival of the surplus than Congressional Budget Office Director Rudy Penner did, by some oversight he fails to correct Penner's explanation of why spending discipline collapsed, in DeLong's words, "at the beginning of 1998 ... so completely."

"So what happened at the beginning of 1998 to change things so completely? That's still not clear to me...
"Rudy: I believe it was the surplus. PAYGO was originally designed to stop tax and entitlement policy from increasing the deficit. (Really, to preserve the gains from the 1990 budget agreement.) After 1997, it had the effect of preventing any reduction of the surplus and that didn't make much sense...."

Achilles should send Rudy an instructional e-mail, he's at the Urban Institute now.

3) In his rush to refute my statement that the early year surpluses that Congress & Clinton so happily took credit for, and used as justification to boost spending (as per above) starting in the 1998 election run-up, resulted entirely from the Social Security surplus (without Krugman saying "peep") he posted a long list of numbers claiming to show the opposite.

Yet he neglected to post these...

1998 Budget surplus $69.2 b, SS surplus $99.0 b
1999 Budget surplus $125.6 b, SS surplus 124.7 b

So we can say that of the $194.8 billion surplus over those two years, merely $223.7 billion came from SS, eh?

And that thus, not counting the SS surplus, there'd have been *no* surplus that the politicians could have taken credit for in the 1998 election run up, politically giving them the free hand to dole out so much more election year pork than in prior recent election years, through the increase in discretionary spending that Achilles kindly has documented for us.

Jim,

If your original claim was "if you average spending increases before the surplus in 1998 and then average the spending increases in the periods after the surplus, including 2000 and 2001, then put a negative number in front of the first and then in an amazing feat of math can call it an 11 times increase" I would say I am wrong and you are right.

You made a very specific statement about spending increasing by 8% a year from 1998 through 2002, and I posted the numbers to say that claim was wrong. I don't know how Morgan Stanley calculates budget deficits and I am not sure I need to use Morgan Stanley numbers instead of CBO numbers.

Second, at no point did I say I know more than Rudy Penner did. I simply posted numbers that contradicted what you posted. So don't pretend that I came across as an expert on the budget. There are two explanations: there is some subtlety to Penner's argument that does not show up in the data (and I offered up the possibility that I was missing something in my post) or most likely you took Penner's argument and added your own 8% a year explosion of spending story to it as spin. [In your post, it did not seem like the 8% a year figure came from Penner.] But hey I was just posting the numbers, not making outlandish claims that are not backed by data.

And on your final point of verbal jujitsu, my data clearly showed that ON budget balances were trending towards smaller deficits. This was to refute your story about how exploding spending was obscured by the social security surplus allowing Clinton to take credit where there was none. The refutation was that

a) Spending was not rising at 8% rates
b) On budget deficits were falling
c) The combination of b) and social security surpluses led the overall budget into surplus.

I am sorry that the facts don't agree with the spin you like to put on the situation.

I agree with several of your posts that you have made here and elsewhere, but not this one.

Okay, I took an unnecessary gratuitious shot there, he did not multiply the negative number.

GT, you do seem to adhere to the theory that politicians and campaign consultants are completely irrational, in that they consistently harp on a issue (deficits) that has zero impact on their electoral propspects. Perhaps you are right. On the other hand, there may be a chance that engaging in deficit spending does carry some political cost, which is why politicians talk about the issue on the campaign stump. Of course, observers can make their own conclusion as to which model is more plausible.

GT, you do seem to adhere to the theory that politicians and campaign consultants are completely irrational, in that they consistently harp on a issue (deficits) that has zero impact on their electoral propspects. Perhaps you are right. On the other hand, there may be a chance that engaging in deficit spending does carry some political cost, which is why politicians talk about the issue on the campaign stump. Of course, observers can make their own conclusion as to which model is more plausible.

Not sure what that has to do with anything we are discussing but sure. Why not?

achilles,

DeLong is definitely stating his preference with: "a decision to award victory to one side". He doesn't like that side, as any casual reader of his blog will recognize (e.g. "why are we ruled by these cretins").

As to Cecchetti's "framework", simple algebra tells us that if you have a revenue cap AND a cap on debt (both ironclad), you've got a spending cap. But, in fact, congress lifts the debt limit when we bump against it.

A little more algebra will show that a reduction in revenue, along with an inviolate debt limit, must limit spending.

Not even "printing money" will overcome it. Even if the money is literally distributed by Milton Friedman's helicopters--the govt will only get it in tax revenues, which, in Cecchetti's model are capped (unlike in the 70s when "printing money" did increase tax revenues through "bracket creep").

And I want top say a word about the disingenuous claims about maximum spending. For forty years federal spending has been higher than 23% of GDP, ONLY TWICE (maximum; 23.5%). With top tax rates as high as 90% and as low as 28%. In inflationary and non-inflationary eras. During wars and peace. That should tell us something.

Well, GT you seem to be asserting that reducing taxes has zero impact on spending since borrowing simply makes up the difference. If this is true, then campaign consultants and candidates are completely irrational in their use of deficit spending as a campaign issue, since, by your reasoning, they obviously don't believe that the borrowing they employ has any actual political cost. In other words, according to your argument, their actual behavior, in regard to borrowing, has nothing to do with what they try to use as useful issues when retaining or obtaining office. So, which is it, do political officeholders believe that they can simply borrow without political cost, or do they believe that deficit spending carries potential political cost, which is why they use it as an issue in campaigns. Or do they talk about deficits in campaigns because they are drooling idiots, and could just as well perform Gregorian chants with as much effect?

Patrick,

I think you and I will have to respectfully disagree on the conclusions we draw from the DeLong quote. I agree with you that a debt cap and a tax cap limit spending: but I still believe that DeLong's argument is that in a time where spending is predicted to grow to a significant fraction of GDP we should be thinking about reform instead of caps (the last two sentences of DeLong's quote).

In the end, what you should do is pose the following questions directly to Brad DeLong.
a) "Do you think the size of government is too small?",
b) "If so, what do you think it SHOULD ideally grow to?"
c) "What do you think the extra spending SHOULD be composed of?"

I would love to know the answers to these question. My prior is that DeLong does not live up to the "big spending liberal" caricature that most conservatives like to attribute to anyone who is left of center. He's not going to be Will Allen, but he's also not going to be a big government type.

Well, GT you seem to be asserting that reducing taxes has zero impact on spending since borrowing simply makes up the difference.

Hmmm. Zero impact? That sounds a bit harsh. I’d say no discernible impact based on the evidence I looked at.

If this is true, then campaign consultants and candidates are completely irrational in their use of deficit spending as a campaign issue, since, by your reasoning, they obviously don't believe that the borrowing they employ has any actual political cost.

You ask an interesting question. But it’s a bit unclear so I’m not sure I can answer it.

But before I do notice how your counterarguments always avoid the data. Instead of saying that my analysis of the data is wrong or that there is other data that proves me wrong you keep coming up with these theoretical constructs. That should tell you something. You are beginning to sound like Max Weber (I believe) who was once told that the data contradicted his theory and said “Too bad for the data”. It may be a made up quote but it sounds about right!

So your new counterargument, as I understand it is that politicians use deficit spending as a campaign issue and therefore…what exactly?

Here’s where I am not sure I follow your point. Are you saying that since politicians talk about deficit spending that means that voters are concerned about it and they can’t simply borrow to finance the debt and therefore they must cut spending or limit it somehow?

If so I would say:

1) Some politicians (including today the Bush administration and the GOP) are arguing that deficits DON’T matter. That directly contradicts your point. It’s true that in the last 2 decades (mainly) Democrats have argued that deficits matter but it’s not clear that voters ever cared about that. They don’t seem to have voted on that issue at all.

2) Your point assumes that if a politician says something that must mean it is relevant to voters. That is nonsense. Politicians push issues that end up not having any real impact all the time. Look at the GOP and impeachment. Or the Dems and union rights in the Homeland security bill. Politicians misjudge all the time.

3) Finally, and most importantly, even if you were right and voters did care about the deficit it does not follow that it need be resolved by limiting spending. It can equally well be solved by raising taxes (Clinton and Bush I).

At the end of the day this remains an empirical question and requires an empirical answer. Surely if the data supports your view you (or Jane) should be able to find SOME research somewhere that shows, say, that cutting taxes by $1 will reduce spending by 20 cents.

Is there no empirical support AT ALL for your thesis?

The data you present answers a question, what happened to spending after a tax cut, while failing to answer the more pertinent question, which is what would spending have been absent a tax cut. Empirical data cannot supply the answer, any more than empirical data can tell us how the U.S. economy would have performed in the years 1941-1946 if Pearl Harbor hadn't been bombed. Pearl Harbor was bombed, and all we have is the data that reflects the U.S. economy while engaged in world war, not data that reflects the U.S. not engaged in World War. Thus how the U.S. would have performed without entry into WWII becomes unknowable, because there are too many complex variables that can't be accounted for. We have no way of knowing what spending would have been absent the tax cut; all we have is the data that tells us what the spending was after the tax cut. This isn't like asking what would have happened to an ice cube if it hadn't been taken out the freezer, it is more like asking what would have happened to a particular water molecule in the Pacific Ocean if a typhoon had not developed somewhere between Japan and the Hawaiian Islands.

This isn't criticism, per se, since the second question is counterfactual, and hence unanswerable. Note that I never claimed to know whether a tax cut would reduce spending, I simply claimed to know that a tax cut would cut spending potential, which I consider a positive good. You replied that taxes collected were irrelevant to spending, since borrowing potential was so huge, which implies that those who make spending decisions believe that there is no political cost to borrowing, and therefore will not be deterred from spending by reduced tax revenues. I merely observed that this is not how such people actually behave, and you kindly noted that many such people have behaved for several years as if borrowing DOES carry a political cost . So is it now your position that nearly every Democratic political candidate of the past 20+ years is an utter moron, by harping, one election cycle after another, about deficit spending, when engaging in deficit spending carries practically zero political cost? Or might it be more plausible to believe that very intelligent people, who have spent hundreds of millions to measure voter sentiment, have concluded by empirical means that engaging in deficit spending does carry some degree of political cost, thereby making it worthwhile to talk about, not as a one-time experiment, but as an ongoing tactic over nearly two dozen election cycles? If an activity has that sort of measurable import, to a large enough segment of voters, over a period of 20+ years, might it just be the teeniest, tiniest, bit plausible that some elected officials might be just the teeniest, tiniest, amount affected? There really is no way around it; for your assertion to hold, nearly every Democratic candidate, and a good portion of Republican candidates, must be utter, complete, drooling imbeciles. Well, I don't care for most of them very much , but give the Devil his due. These positions are highly coveted by extraordinarily hard working people, who desire to gain access to power for a variety of reasons. The chance that a majority, or even a a large percentage of such people are morons is practically nil.

Maybe we don’t disagree that much. You seem to be focused on ‘spending potential’ and I am focused on what actually happened.

Given how much room we still have to raise and finance spending I don’t find the concept of spending potential analytically useful. We could easily double our spending if we needed to (as we did in ww2). To talk about reducing our spending potential is like talking about reducing the number of hamburgers one can eat from 200 a day to 180 a day and saying that was good to reduce obesity. It is so much higher than the actual numbers as to be meaningless.

I fail to understand though, your election arguments since Republicans are saying that deficits don’t matter. Presumably they also do research yet it’s not telling them what you think it should. And Democrats, who do say it matters, are not proposing to cut spending. So neither of the two supports your argument.

At the end of the day I think that the reason why spending is disassociated from taxes in the data is because they are two separate things in voters minds. That’s why they will ask for tax cuts but not support any meaningful spending cuts. Our huge financing ability allows the government to deal with these two almost opposing views by issuing debt when needed. Recent history shows that when the difference becomes too big taxes are raised again, and spending is not cut.

My personal prediction? Spending will never go back to the pre-welfare state levels so many conservatives wish for. In fact demographics will soon conspire to raise spending even further and I would not be surprised to see govt spending at 25% of GDP in a decade or so. Cutting taxes will not make one iota of a difference unless you change voter’s views of what they want. So enjoy your tax cut while you can. It probably won't last.

Achillies, perhaps I am misinterpreting you, but is it your assertion that national government has an infinite ability to borrow; that it could endlessly finance it's activities soley through borrowing? If not, then it logically follows that reducing taxes reduces the ability of government to spend money.

My assumptions:

1. Inflation is a tax on holding money.
2. Issuing debt is a tax on investment.
3. A tax on incomes is just that.
4. Government spending, as a mathematical identity, is paid for by resources diverted from the private sector, either by things called "taxes" or things which are in effect taxes (inflation, debt).

My conclusion: deficit or inflation-financed tax cuts do not change the tax burden; they redistribute it.

Interestingly, this writes the "deficits make government spending untenable" theory as "deficits shift the tax burden to those who hold investments"; that is, the rich. So, basically, what you've got is:

Goal: Lower the tax burden on the rich.
Requirement for this: Cut government spending.
Method for cutting government spending: shift the tax burden to the rich. This will make the rich so angry that they'll cut spending!

Hilarious. Reminds me of the green party/communist theories on "heightening the contradictions" or "things have to get really bad first" to make the revolution come.

Jason, why is it more anti-democratic to obtain a vote by promising to allow the voter to keep more of what they have obtained via mutual agreement, while failing to address what borrowing might thereby be required, as opposed to obtaining a vote by promising to forcibly take property from another citizen, and give it to the voter, while consistently underestimating the amount of taking that will be required to fufill the promise?

Can you rephrase this? I've read it 5 times and can't figure out what it has to do with control of the government through voting (democracy).

Achilles still doesn’t post the relevant numbers:

>>You made a very specific statement about spending increasing by 8% a year from 1998 through 2002, and I posted the numbers to say that claim was wrong

You never posted numbers for 1998-2002. Here they are: Nondefense discretionary spending ...

1998 281.9 billion
2002 392.5 billion
http://www.whitehouse.gov/omb/budget/fy2003/pdf/hist.pdf, Table 8.1

That’s a 39.2% increase over four years, and if you get out your calculator you’ll find it compounds to 8.6% per year.

Now I hope that’s settled. As for...

>>My data clearly showed that ON budget balances were trending towards smaller deficits...

Well, golly. ;-)

>> This was to refute your story about how exploding spending was obscured by the social security surplus allowing Clinton to take credit where there was none.

The budget data I posted and that you fail to refer back to shows, just as I said, that there was an operating budget *deficit* in the 1998 election year, excluding Social Security.

Nevertheless, all that year Bill and Newt claimed credit for a surplus in preparation for the election, and that was the year spending discipline broke down “due to the surplus”, as per the description of CBO Director Penner, and as per the zooming up of the spending numbers published by both you and me.

Now as it’s hard to argue with those facts I don’t quite feel “refuted”.

In fact, for anyone who doubts that government spending follows revenue upward, this is about the most blatantly obvious example one could hope for -– until the even better one came along of California in particular and the other states in general building in all that temporary boom revenue into permanent budget lines.

But back to the Feds: During the deficit years discretionary spending grew *less than inflation*. Then when the surplus hit it accelerated every year -– up 13%(!) in 2002. What better example could one want?

BTW, now that deficits are back, discretionary spending is scheduled to grow only 4% next year, in the latest budget proposal -- heading down again.

Okay this is getting old, and I am tired of moving targets. I was not hiding anything, my data was the same as yours, except for two things:

1) I used DOMESTIC non-defense discretionary spending instead of overall non-defense discretionary spending (does not make much of a difference: I focused on domestic for the reason that politicans get utility from domestic pork)

2) More importantly, I only used data through 2001, since the 2002 were only estimates. Jim uses the 2002 estimated data, which turns out to be a large part of the story.

So for the last time here are the numbers, Jim's preferred metric non-defense overall discretionary spending in billions of NOMINAL dollars.

Year Level Change from previous year
1997 275.6 3.34%
1998 281.9 2.29%
1999 296.5 5.18%
2000 319.9 7.89%
2001 348.3 7.31%
2002 392.5 14.32% (!!!!!!!)

Notice how crucial the inclusion of 2002 is to get the 8% number that Jim claims. And by 2002, the on-budget surplus was 260 billion in the red, no one was expecting surpluses at any point in the upcoming three years, yet we are supposed to believe this is Congress spending bloated revenues?

As I said in my last email, your story was that spending rose by 8% a year when surpluses appeared is not true. The average comes out to 8% a year, but the average is driven by two large numbers that appeared in the two years after the surplus was long gone.

Hey, but that is just like the $1000 average tax cut, right? A couple of small numbers added together with a couple of large numbers and a dashing tale and who cares about what reality has to offer.

Finally let me close with the observation that in 1998, the year that Jim argues "spending discipline broke down" the on-budget balance went from -103 billion to -30 billion. In 1999, the year after discipline broke down, the on budget balance went from -30 bilion to +2 billion and 2000, two years after discipline had allegedly broken down, the on budget balance went from +2 billion to +86 billion. Now you have to live in a world of pretty remarkable lags to claim that discipline had broken down and we were burning tax revenues like crazy. But, why let the facts get in the way of the nice story.