It is important that the deficit is first framed correctly and to get past all of the politics of it. What it boils down to is that this is an issue that primarily relates to revenue, rather than expenditures. Yes, one could argue that if we reduced our expenditures to a more "reasonable" level (whatever that means), we could make current revenues adequate. However, in terms of actually determining how we got from moderate, though still large deficits, to extraordinarily huge ones, we need to look at the revenue side of the ledger. What we find here is that, despite these assertions by some that taxes are far too high, taxes as a percentage of GDP are the lowest the have been in recent memory.
What we have seen since the recession began is a large one time spike in expenditures due to higher automatic stabilizers (unemployment benefits, food stamps, Medicaid, etc) and of course the bailouts, which were necessary to prevent a complete meltdown. Unfortunately, the aspect of this that isn't one-time is the large addition of debt, which will increase debt service payments dramatically. Most of the other increased expenditures enacted over the past few years are nothing out of the ordinary and don't amount to a dramatic increase in permanent spending much beyond what would have occurred otherwise.
The combination of the 2001 and 2003 tax cuts along with the cut punch of this recession has driven tax revenues as a percentage of GDP to right around 15%. With an economic recovery and the reversal of the Bush tax cuts, the CBO projects that revenues would return to over 20% of GDP by fiscal year 2014. Of course, the difficulty is that there is not a chance we will be able to reverse all of the tax cuts, even though we really should do so. As a consequence, instead of a 3.0% of GDP deficit by the middle of the decade, we are probably looking at more along the lines of 4-4.5% of GDP, depending on how much of the tax cuts we retain.
We do need to recognize that we need more than just tax increases on upper income earners in order to solve the revenue side of the equation. There needs to be a broad based tax increase, possibly led by what the deficit commission has laid out with the elimination of a variety of tax expenditures including the mortgage interest deduction. I am particularly in favor of that since I've believed for years that the mortgage interest deduction is one of the many things that drastically distorts the housing market. I would be in favor of keeping it in a compromise, but with a low maximum deduction. The tax code, generally, needs to be simplified so that the rates are more meaningful both to taxpayers and to policymakers, but this needs to be done in a decidedly revenue additive way.
Additionally, while a progressive tax code is desirable, we do need to increase taxes in a broad way since relying on only the highest tier of income produces a great deal of revenue volatility as bonus and interest income that is subject to the highest marginal tax rate is immensely variable in economic booms and recessions. One way of achieving a broad based tax increase that would be gradual would be to suspend indexing of the brackets for several years while also raising rates on the top three brackets. In an ideal world, income between $100,000 and $200,000 would be taxed at 35%, between $200,000 and $350,000 at 40% and $350,000 and above at 45%. There should be no offsetting tax cuts whatsoever. The only shift I think would be desirable would be the elimination of the corporate income tax and a commensurate shift of burden onto personal income taxes. Of course, good luck getting that done politically.
Though I've generally been in favor of preferential tax rates for investment income, when you look at the national income accounts, it is clear that we can't just wall off dividend income from paying its fair share of taxes. We're talking around $800 billion to $1 trillion a year in dividend income and, additionally, several hundred billion a year in capital gains realizations. Further, due to various tax schemes that we can get into another time, a lot of income gets sheltered into the more favorable investment tax rates (see carried interest). In more flush times, I might be in favor of maintaining the preferential treatment of investment income, but as it stands now this is a sacrifice that might have to be made. A consequence would be a contraction in valuations since expected after-tax earnings on investments would be reduced.
I haven't done all of the math, but instead of what the deficit commission has suggested with taxes as a percentage of GDP never topping 21%, we should approach it with a mind to increase taxes as a % of GDP to 23% or even 24%. It's an inevitable payback for having taxes far too low in relation to spending for too long. It's not that spending has been out of control, but rather that we haven't been paying for our existing commitments. As a consequence, the deficits accumulated have rolled up into a large future debt service burden that will crowd out other necessary areas of federal spending that are realistically too painful to cut back on.
On the expenditure side, in the short run defense spending has to bear the brunt of the reductions. Outside of one-time economic stimulus spending, non-defense discretionary spending has not increased a great deal in a fairly long time. That's not to say that it shouldn't be reduced, but that quite frankly there isn't enough there to cut. In defense spending, there is a good deal of froth that doesn't do us very much good. I know some have tried to argue that defense spending is more stimulative than other forms of government spending, but that is utterly preposterous. If someone can tell me how this:
I'm all in favor of a non-defense discretionary spending freeze starting in FY13 and possibly even reducing spending then by 5%. In truth, most of the effects here are marginal at best, but there is some effect there to be sure. I'm not willing to cut NIH and some of the other major chunks of the non-defense discretionary budget that are vital to long term research and development in this country, which is where we need to be focusing investment. This economy will only continue to be competitive with massive investments in human and physical capital and those have to be well targeted. Spending cuts must be prioritized based on economic impact. Those with less economic disruption must bear more of the burden than those that have high returns in the sectors that are our growth engines in science and technology.
As far as the timing of all of this, it is not vital nor prudent that any of this be done right away. It is not vital because the deficit will narrow considerably as some of the one-time revenue hits and spending hikes wear off. It is not prudent because a sudden and drastic fiscal contraction of 3-4% of GDP would invariably cause a recession and once again depress tax revenues into the 15-16% of GDP region. In the next fiscal year we face a steep contraction in state and local government budgets that, combined with severe federal budget reductions, could drastically damage domestic consumption. We need to get over the hump and be well on the way to sustained economic growth before taking on a sharp fiscal contraction. If we repeat the mistakes of 1937-38, we will regret it. In general, I would prefer to wait at least one year before bringing on the pain. Some of the defense cuts can be done sooner than that since their effect on the macro economy is likely to be quite muted. The mix of the fiscal contraction should be initially weighted toward high income tax increases and defense spending cuts since those have the smallest short term contractionary effects on the economy.
While Social Security and Medicare changes are not that important in the short run to righting the deficit, in the long term the deficit will become truly massive nearly entirely on account of Medicare. Social Security is easy to fix where you eliminate the cap on taxable earnings and also go to lower inflation factors for benefits. This would actually throw off a lot of excess cash that would inevitably go to the general fund. What's better is that neither change would undermine the basic functions of the program. These changes can be implemented years from now and we will still avoid disaster.
Medicare is tricky as the policy levers available are either brutally simple (strictly capping benefits, raising eligibility age, etc) or nearly impossible to assess (increasing competition in the health care industry, capping medical malpractice awards, better enforcement actions). I generally subscribe to the view that at some point health care inflation will slow down considerably, which is not baked into most of the long term budget forecasts. Regardless, CBO has modeled what happens under such slowdowns and you still come nowhere near closing the gap. I have to admit I don't have good answers here beyond draconian ones. I am willing to wager that health care inflation won't always be such a high multiple of inflation in general solely because at the rates the "experts" are speaking of it would crowd out the rest of the economy. At some point before then, employers would scream and drop coverage if costs didn't come down and state and local governments might do the same. Even so, the aging of the population, which is a somewhat (though not entirely) separate issue from high rates of health care inflation, will drive up expenditures significantly as is. This simply cannot be bridged with tax increases because the sorts of tax increases we would need would be utterly ruinous. Unless a strong regimen of cost control is implemented in one form or another, this will destroy us. To repeat, I don't have good answers here.
Now, let me briefly address those who think that you can cut taxes and the deficit at the same time. This is baloney. The reason why tax cuts generally are followed with higher nominal dollars of revenue is that the tax cuts that have been implemented are smaller than the underlying growth in revenue. There is basic simple math that makes it nearly impossible for tax cuts to pay for themselves. If you are at 20% of GDP in taxes and you implement 1% of GDP in tax cuts, GDP has to grow 5% above where it would have in absence of the tax cuts in order to make up the gap you have created in revenue. That's a multiplier of 5x. I've never seen a single study that suggests anything like that is even remotely possible and most suggest multipliers of less than 1x. It's a little more complicated than that because the distribution of income might change more heavily toward upper income groups with lower rates as upper income earners might be comparatively more incentivized to work, pushing more income into higher marginal rates. Even so, the numbers are
Those are my thoughts for now. I regret not having more specific dollar amounts available since I didn't have the time necessary to model all of this even in a rough fashion. Luckily there is a New York Times tool that has some estimates for various proposals: https://www.nytimes.com/interactive/2010/11/13/weekinreview/deficits-graphic.html