I am about to venture slightly out of my usual comfort zone, monetary theory, history, and policy to talk about taxes. You guessed it, March is almost gone, and I still haven’t filed my income tax returns. But that’s not what I am going to write about. I am going to discuss an article by Eduardo Porter (“The Case for Raising Top Tax Rates’) in today’s New York Times Business Section which caught my eye. Porter gives a good summary of the sea change in tax policy that was inspired by a brash young economist with a Ph. D. from the University of Chicago working in the Nixon and then Ford Administration in the early 1970s. His name was Arthur Laffer. Porter tells the (apocryphal?) story of how Laffer drew his immortal curve for his bosses (Donald Rumsfeld and Dick Cheney — once upon a time able public servants!) on a cocktail napkin showing that any feasible amount of tax revenue could be generated by two tax rates (one high and one low) except for the maximum total revenue achievable only by a unique rate. The Laffer curve became the inspiration for what became known as supply-side economics. Despite enduring much ridicule, the Laffer curve became the central plank in the platform on which Ronald Reagan won the Presidency in 1980. Cutting taxes became the unifying principle on which almost all Republicans could agree, becoming the central pillar of conservative and eventually Republican orthodoxy. It was not always so. Barry Goldwater voted against the Kennedy across-the-board tax cuts of 1963. His vote against tax cuts, cuts supported by his chief opponent for the Republican Presidential nomination in 1964, Nelson Rockefeller — the object of conservative revulsion and outrage to this day — did not evoke so much as a peep of protest by conservatives when Goldwater was carrying the conservative torch in his epic campaign for the GOP nomination.
But as Porter points out, it’s not just conservative and Republican views on taxes that have changed. There was bipartisan support for cutting rates in 1986 when the top marginal rate was cut to 28%. Although Democrats consistently want to raise the top marginal rate, President Obama has not proposed raising the top marginal rate even to 40%, 10% below the 50% top rate under the Reagan tax cuts of 1981.
Why this sea change since the 1960s in views about top marginal rates? Obviously, tax cutting has proved itself to be politically popular, and that may be all the explanation necessary to account for the change in the political consensus about how high marginal tax rates can be raised. But Porter also notes that there was an important economic component in support for keeping tax rates low.
For 30 years, any proposal to raise taxes had to overcome an unshakable belief that higher taxes inevitably led to less growth. The belief survived the Clinton administration, when taxes rose and the economy surged. It survived George W. Bush’s administration, when taxes were cut yet growth sagged.
Porter cites new research suggesting that the scope for tax increases is really a lot greater than had been thought. Not only could the government “raise much more tax revenue by sharply raising the top tax rates paid by the richest Americans, but it could do so without slowing economic growth.” To support this idea, Porter cites a recently published paper by Emmanuel Saez and Nobel Laureate Peter Diamond. They argue that raising the top marginal rate to 80% would NOT cause revenue to fall if the loopholes available to the rich were closed. One obvious problem with that proposal is that the rich have a huge incentive to spend money lobbying against any increase in rates that is accompanied by closing loopholes and against any closing of loopholes not accompanied by a reduction in rates. Guess what? Spending money on Congress works, so don’t hold your breath waiting for tax rates to rise while loopholes are closed. But Saez and Diamond also estimate that even without closing loopholes the top marginal rate could be increased to 48% before any further rate increases would reduce revenue.
Moreover, in another study Saez, Slemrod and Giertz found that although the rich would respond to increases in marginal rates by trying to shelter more of their income, doing so would not cause economic growth to slow down. Porter explains:
That’s because a lot of what the rich do does not, in fact, generate economic growth. So if they reduced their effort in response to higher taxes, the economy wouldn’t suffer.
Porter adds:
The arguments are not the mainstream view. Some economists really dislike them. And they are not absolutely airtight. The calculations rely on estimates about how higher tax rates would discourage the rich from working or investing over a couple of years at most. But we know little about how they might affect long-term decisions, like whether to become a brain surgeon or a hedge fund manager. We do know that in countries with higher tax rates, like France, people work fewer hours than in the United States.
Is there any way of explaining why raising top marginal rates to very high levels would not cause a loss of real income? Here’s an idea. The era of low marginal tax rates in the US has been associated with a huge expansion in the US financial sector. Wall Street has consistently been paying the highest salaries and giving the largest bonuses ever since the 1980s, attracting the best and the brightest from each year’s new crop of grads from our elite colleges and universities. What has been the social payoff to this expansion of finance? I am not so sure. Over a century ago, Thorstein Veblen wrote his book The Theory of the Leisure Class, followed some years later by his essay “The Engineers and the Price System.” He distinguished between engineers who actually make things that people use and financiers who simply make investments on behalf of the leisure class, adding no value to society. This was a vulgar distinction, premised on the unwarranted assumption that finance is unproductive simply because it generate no tangible physical product. On that criterion, Veblen would have ranked pretty low as a contributor to social welfare. Mainstream economists felt pretty comfortable dismissing Veblen because he was presuming that only physical stuff can be valuable.
However in 1971, Jack Hirshleifer, one of my great teachers at UCLA, wrote a classic article “The Private and Social Value of Information and the Reward to Inventive Activity.” The great insight of that article is that the private value of information, say, about what the weather will be tomorrow, is greater than its value to society. The reason is that if I know that it will rain tomorrow, I can go out today and buy lots of cheap umbrellas (suppose I live in Dallas during a drought), and then sell them all tomorrow at a much higher price than I paid for them. The example does not depend on my having a monopoly in umbrellas; I sell every umbrella that I have at the rainy-day market price for umbrellas instead of the sunny-day price. The gain to me from getting that information exceeds the gain to society, because part of my gain comes at the expense of everyone who sold me an umbrella at the sunny-day price but would not have sold to me yesterday had they known that it would rain today.
Our current overblown financial sector is largely built on people hunting, scrounging, doing whatever they possibly can, to obtain any scrap of useful information — useful, that is for anticipating a price movement that can be traded on. But the net value to society from all the resources expended on that feverish, obsessive, compulsive, all-consuming search for information is close to zero (not exactly zero, but close to zero), because the gains from obtaining slightly better information are mainly obtained at some other trader’s expense. There is a net gain to society from faster adjustment of prices to their equilibrium levels, and there is a gain from the increased market liquidity resulting from increased trading generated by the acquisition of new information. But those gains are second-order compared to gains that merely reflect someone else’s losses. That’s why there is clearly overinvestment — perhaps massive overinvestment — in the mad quest for information.
So I am inclined to conjecture that over the last 30 years, reductions in top marginal tax rates may have provided a huge incentive to expand the financial services industry. The increasing importance of finance also seems to have been a significant factor in the increasing inequality in income distribution observed over the same period. But the net gain to society from an expanding financial sector has been minimal, resources devoted to finance being resources denied to activities that produce positive net returns to society. So if my conjecture is right — and I am not at all confident that it is, but if it is – then raising marginal tax rates could actually increase economic growth by inducing the financial sector and its evil twin the gaming sector — to release resources now being employed without generating any net social benefit.
UPDATE 9:39PM EDST: I left out “not” (now in CAPS) in between “would” and “cause to fall” in the fourth sentence of the paragraph beginning with “Porter cites.”
UPDATE 9:46PM EDST: I revised the final sentence of the same paragraph which had been unclear.
Uneasy Money 
While I can’t support raising marginal tax rates in order to discourage arbitrage, I have been thinking along these lines regarding the payments system. How much effort has been made to allow huge and rapid finanicial transactions? For example, a trader needs to buy $700,000,000 with of some financial asset this morning and they sell it all this afternoon? How buy now and sell in 20 minutes? And, “we” need to make sure that this generates to systematic risk. That is, the person selling to this trader have to be able to use the receipts to make a huge purchase in a few minutes, and the person selling to them should bear little or no risk that the funds are not good.
If we made everyone cart around suitcases full of cash to make these payments, they would all be sure, but could it be done?
Now, how important is it for people to be able to make all of these transactions?
Suppose Microsoft wants to build a plant and needs $300,000,000. The investment banker who wins the contract ends up wiring Microsoft $300,000,000. Microsoft just leaves the money in the bank, or more realistically, buys short term to maturirty securities of various sorts. And then, as they actually contruct the building, they sell of the securities and make payments.
But the investment bank had to raise $300,000,000 right away. They likely sold a bunch of short term securities, to raise funds to be wired to Microsoft.
In theory, the investment bank will sell the microsoft stocks or bonds, and pay off the short term funds they borrowed so they could give microsoft all the money.
What if the investment bank must gave microsoft the money gradually as they sold of the stocks and bonds? What great efficiency gain is their in having the investment bank borrow money short and give it to microsoft who will just lend it short?
And this may seem a bit crazy, but to what degree is this an artifact of interest rate targeting?
LikeLike
John Bogle (founder of Vanguard Mutual Funds) has been making a similar argument for a while now. His book “The Battle for the Soul of Capitalism” touches on the ZMP of the finance sector construed broadly.
James Livingston, history professor at Rutgers, makes a corollary argument to yours in his essay “Their Depression and Ours”. http://hnn.us/articles/55614.html Crudely summarized, he argues that in the 1920s and today, that there was a massive shift of income shares to profits, away from wages and consumption. Profits produced surplus capital that was invested speculatively – in stocks then, housing derivatives today – which produced bubbles that would necessarily prove unsustainable. All this occurred when the production of consumer durables and consumption of those durables and service became the main driver of economic growth.
I think you’re on to something here. I’m interested in how “soaking the rich” might have produced more economic stability, as opposed to the low marginal rates regimes which produces large profits that get invested speculatively, producing bubbles.
LikeLike
Interesting post.
I wonder if lower top marginal tax rates have contributed to the global explosion in capital we have enjoyed in last 20 years. The upper classes can afford to save, and so they do (even after conspicuous consumption).
If the capital can be harnessed to productive means, maybe it is a Faustian bargain.
BTW, the top federal income tax rate in the 1950s through the 1960s was 90 percent! Rough-house guess, you had to make $10 mil (today’s dollars) to get into that bracket.
There is an original Twilight Zone episode out there (c. 1962?) about a couple who get a million dollars from a genie, go on a spending spree, but then find out they owe $900,000 to the IRS. After paying taxes and the spree, they have nothing again.
They decide money was not important after all! Imagine someone paying 90 percent in taxes today and then saying “la-de-da.”
But we had good growth in the 1950s and 1960s, with that 90 percent top rate. And not much inflation. And we were paying down an onerous national debt incurred in WWII.
LikeLike
“The Laffer curve became the inspiration for what became known as supply-side economics. Despite enduring much ridicule, the Laffer curve became the central plank in the platform on which Ronald Reagan won the Presidency in 1980.”
The Laffer curve is a revenue maximization curve. What Laffer never addresses is why revenue maximization is important.
Prior to 1980, the bond market would respond to higher inflation with higher interest rates. That was because most of the federal debt was held by pension funds, insurance companies, and the like who actually cared about the amount of interest they were receiving (unlike our trade partners who buy our debt for other reasons).
And so, revenue maximization is a means to limit the federal deficit and debt.
Why is limiting the federal debt important?
Prior to 1980, the “crowd out effect” theory was in vogue. It is a situation that can occur in closed economic systems. It basically says that under a fixed money supply, federal government borrowing tends to crowd out other forms of borrowing. A lot of it is nonsense since the money supply is not fixed and we don’t have a closed economic system but at the time guys like Milton Friedman were arguing that the money supply should be regulated / constrained and so there you have it.
The crowd out effect does not exist in economic systems that permit the free flow of capital. Instead what happens is that monetary policy loses its ability to differentiate between productive and non-productive enterprise (not that monetary policy alone can do this).
It would seem Mr. Laffer should go back to the underpinnings of his curve, and realize how far off base the current crop of Republicans are (Federal deficit – +$1 trillion annually, federal debt +$10 trillion and counting).
Unless of course you realize the whole problem with Mr. Laffer’s approach.
1. Revenue maximization is not important because the federal government literally cannot go bankrupt. And the reason the federal government cannot go bankrupt is that it’s revenue stream is a legal obligation rather than a discretionary purchase and that payments on its bonds supersede all other payments courtesy of the 14th amendment to the Constitution.
2. The federal government is not required to sell debt to finance its deficits any more than General Electric, or Microsoft, or Goldman Sachs has to sell debt to finance theirs. The federal government could just as easily sell equity like claims on its revenue.
But of course none of this makes it through the thick skulls of our current Congress men and women (approval rating around 12% – lower than Nixon’s rating during Watergate).
LikeLike
John Bogle (founder of Vanguard Mutal Funds) has been making argument that the financial sector construed broadly is close to zero marginal product. This is touched on in his book “The Battle for the Soul of Capitalism”.
James Livingston has also touched on this in his essay “Their Depression and Ours” which compares the Great Depression to today. http://hnn.us/articles/55614.html He argues that in the 1920s company’s income shifted away from wages into profits, and those profits were speculatively invested into bubbles that were bound to end badly. He doesn’t give enough credit to the New Deal monetary policies in generating recovery, but it’s an interesting corollary to the massive rise in income inequality today.
LikeLike
David, you make some interesting points, but I think some are too broad brush.
Consider the financial sector, for example. It’s a huge business that really needs to be broken out by sector: commercial finance, investment banking, retail/mortgage finance, wealth management, insurance, annuities, trading / market making, etc.
I agree that certain aspects of trading, particularly of the high frequency variety and especially flash trading, are harmful because of front-running aspect. A lot of that is incentivized by the regulatory structure rather than marginal tax rate however. Then the question is what other areas of finance are “bad” and too transactional. People are living longer and retiring longer, hence more life insurance and wealth management.
Then I would question how much the “mad rush for information” is a problem with the finance sector rather than with the technology sector. Think google and facebook. Both businesses are predicated entirely on the collection of personal information in order to generate clicks before someone else does. And they do so in an offensive way. For example I repeatedly get ads imploring me to buy gold because I’ve done so many searches for “inflation targeting.” Throw in the penny stock ads, Canadian pharmacy ads, and various other crapola I’ve seen and I’m not sure what value these companies are providing.
At a personal level I do recognize I am a person who could have been a scientist but went into finance instead. So the argument that there might have been better uses for my abilities isn’t lost on me. On the other hand, I think I am pretty insightful about economics, and my observation over the past decade or so is that we live in a world that is pretty lacking in insightful economists. So perhaps there is more bang-for-the-buck if I can constructively add to the wisdom in this field.
LikeLike
Steve:
Your statement:
“I agree that certain aspects of trading, particularly of the high frequency variety and especially flash trading, are harmful because of front-running aspect. A lot of that is incentivized by the regulatory structure rather than marginal tax rate however.”
High frequency trading is encouraged because of the nature of the credit markets, it is cheaper to borrow short term than it is to borrow long term. Tax policy can play a role here by making it cheaper to borrow long term than it is to borrow short term – and it has nothing to do with marginal tax rates.
LikeLike
I’m pretty sure Laffer’s Ph.D. is from Stanford.
LikeLike
You may be right. He was probably an assistant professor at Chicago but not a grad student there.
LikeLike
Enjoyed the article. An example in your favor: the construction of a more direct fiber cable from NYC to Chicago in order to save 20-30 microseconds for HFTs for around $300mm and talk of a similar venture from London (Europe) to Tokyo for five times that amount. I am sure there are sound business reasons for the construction and use of such networks but on a society level a definition of insanity?
Disclaimer: I work in Finance and know some of the people engaged in this practice. It is safe to say they see the cost of this fiber arms race as well.
LikeLike
As someone who trades stock a fair bit, I can tell you that at least 50% of my marketable buys get executed at a price of $xx.xx99 and 50% of sells at $xx.xx01
What does this mean? It means that my orders are getting matched against an HFT book which is offering me 1/100 of a penny price improvement (YAY!). Multiply that by a couple hundred shares and I have… a couple of extra pennies.
Here’s the insidious part: Those pennies are price improvement on marketable orders. But my *limit* orders NEVER get executed unless there is a significant adverse price movement, costing me many dollars. That’s because there is always an HFT ready to jump a millisecond and 1/100 of a penny ahead of my limits when order flow hits the market, unless they determine my limit to be quite bad.
This has everything to do with a market structure that favors exchange fragmentation, millisecond trading, sub-decimal pricing (but only for some people), and in many cases unequal access.
It has nothing to do with marginal tax rates, nor low interest rates.
LikeLike
Seeing that investing in stocks and bonds is old man’s game might not an aging population be a cause for the growth in finance?
BTW some of it seems to of negative value as the money managers overall loose to market in indexes.
LikeLike
BTW, the top federal income tax rate in the 1950s through the 1960s was 90 percent! Rough-house guess, you had to make $10 mil (today’s dollars) to get into that bracket.
When I read something like that my mind always wanders to NBA players like Alvin Iverson. Despite having earned about $200 million, he seems to have been unable to buy some jewelry because he is out of cash (fear not the NBA has held some money back for him). How would he feel if he had been taxed at 90%. Of course he was not investing all his money with Goldman Sachs.
LikeLike
Follow-up questions:
– Do we know for sure that the majority of finance rests on scrounging for information and doing price arbitrage? I think we need data on this.
Finance has done other things too and invented very many useful products in the past 30 years.
– Has the net gain to society really been minimal?
Transacting with a high frequency trader may bring no net benefit to the average Joe, but surely the ability of firms to raise billions of dollars at low costs all over the world, the proliferation of insurance and investment products (I’m happy to have low-cost ETFs that let me hedge gas prices at the pump), etc. are all valuable to society.
Again, I don’t think trading is the only thing finance has done. Trading may just have been a necessary outgrowth of product innovation.
– Finally, the Top 1% does not consistent only of hedge fund managers.
I would strongly argue that the lawyers and doctors that make up big swaths of the Top 1% exploit information asymmetries and barriers to market entry.
This last point is not inconsistent with your broader argument- that the Top 1% are not necessarily doing things in society’s best interests- but I think it’s worth pointing out in light of the concerns regarding finance.
LikeLike
Steve,
You mentioned high frequency traders (HFT’s) being able to front run prices on equity purchases for limit orders. In a zero leverage world, this would be impossible because for every stock that an HFT buys, he / she would have to sell something else and so the net effect on the equity market as a whole would be 0.
Add leverage (borrowed money) to the mix, and suddenly the HFT must be sure that the value of the asset bought has a higher value than the cost of the leverage.
Make borrowing short term expensive and long term inexpensive, and the rules of the game change. Suddenly, it makes sense to hold onto the asset longer and wait, rather than trying to time the market.
And you are correct that marginal tax rates have little to do with this, since they are a fixed cost rather than a floating cost.
LikeLike
Here is a paper that looks at marginal and average tax rates during the past based upon typical “real” incomes in the U.S.
http://journal.apee.org/index.php?title=Fall2009_10
LikeLike
Frank,
If you are doing 10 inventory turns a day, you don’t need very much capital. That’s why I don’t see interest rates mattering.
LikeLike
There is always going to be an inequality in a society that life under capitalism. It is disturbing to know that about 1% of the population holds most of the wealth. This certainly makes me want to become rich.
LikeLike
Steve,
What you mean to say is that in doing 10 inventory turns a day, you don’t need capital for a very long time. Remember the time cost of money? That is what interest rates are. An HFT requires capital (money) to purchase stocks just like everyone else, except that in doing high frequency trades, the time between buying and selling the stock is measured in hours (minutes?) instead of days or weeks. And so an HFT is borrowing money on an hourly basis (the very short end of the yield curve).
This is why I would recommend reversing the borrowing situation through a combination of monetary and tax policy. Increase the cost of short term borrowing through monetary policy, lower the cost of long term borrowing through tax policy.
The beginnings of this are already written into the tax code – short term capital gains versus long term capital gains. Why Congress decided the break point between the two should be 1 year is beyond me.
LikeLike
Frank,
If you have $10 million in capital and ten turns per trading day, you can do $25 billion in trades per year without borrowing anything at all. If you can scalp a measly 2 basis points per trade, you earn $5 million per year. Are you suggesting 50% short term interest rate? For the life of me I have no idea what you are suggesting.
LikeLike
Steve,
What I am suggesting is to change the incentive structure of the credit markets through tax policy, nothing more, nothing less.
Because of a positively sloped yield curve, the short term cost of debt is always lower than the long term cost (pre-tax).
You seem to disregard any possibility that borrowed money is in play in the equity markets.
http://www.nyxdata.com/nysedata/asp/factbook/viewer_edition.asp?mode=table&key=3153&category=8
Total NYSE margin debt: $289.4 billion on a market cap of about $14 trillion.
I don’t doubt that what you say is possible – rolling $10 million of equity 2600 times a year and never selling at a loss is certainly possible however unlikely. And it becomes more unlikely as marginal stock buyers are pushed toward a buy and hold position (high short term cost of margin, low long term cost of margin).
LikeLike
Frank, I initially thought I was having a discussion with a reasonable person, but I’m not sure any more.
You wrote: “What I am suggesting is to change the incentive structure of the credit markets through tax policy, nothing more, nothing less.”
Yes, but you haven’t bothered to articulate a specific tax policy and incentive that you want.
You wrote: “Because of a positively sloped yield curve, the short term cost of debt is always lower than the long term cost (pre-tax).”
Are you aware that you are stating a definition? Why “because” and “always”?
You wrote: “You seem to disregard any possibility that borrowed money is in play in the equity markets.”
We were discussing high frequency trading, not the equity market in general. Please tell me where I said borrowed money was not in play at all and please cut out that attitude if you are going to try to put words in my mouth.
You wrote: “rolling $10 million of equity 2600 times a year and never selling at a loss is certainly possible however unlikely.”
I never said never selling at a loss. The 2bp is a hypothetical average profit on a high frequency trade.
LikeLike
Steve,
“Yes, but you haven’t bothered to articulate a specific tax policy and incentive that you want.”
Okay. The federal open market committee buys and sells short term federal debt to set short term interest rates. The federal debt (unlike a lot of private debt) is not a claim on any physical asset. It is simply a guaranteed claim (by the 14th amendment) on future tax revenue. The federal government could just as easily sell non-guaranteed claims (similar to equities) on the same revenue stream.
What does that mean? Imagine if the U. S. Treasury department told you they were going to sell you an asset that appreciates at say 10% annually over 30 years, but the only way you could realize that 10% was by having a tax liability 30 years from now equal to or greater than the future value of the asset. Lets call the asset a “tax receipt”.
Notice that this is different from a government bond. Your return on investment is conditional on having a taxable revenue stream (wages, sale of assets, etc.) at the time of maturity.
Now suppose you borrow money for 30 years at say 5% at the same time you buy the “tax receipt”. What is your after tax cost of debt for 30 years? Depending on how you are leveraged (future value of debt / future value of tax receipt), your cost of debt could in fact be less than 0% on an after tax basis.
“We were discussing high frequency trading, not the equity market in general. Please tell me where I said borrowed money was not in play at all and please cut out that attitude if you are going to try to put words in my mouth.”
I apologize. I assume we can agree that for every buy and sell order that a high frequency trader makes, there is another trader / investor that is filling the opposite side of the transaction. That other player can be leveraged or non leveraged. If leveraged, then interest rates in part determine his / her investing choices.
“I never said never selling at a loss. The 2bp is a hypothetical average profit on a high frequency trade.”
I guess here is where I am a little lost. The high frequency trader is still buying and selling at market prices with a cushion of 2bp on either side – correct? If so, then the high frequency trader is still subject to those market prices, yes? And so, high market volatility (large intraday price swings) or low market volume (fewer trades) can upset the high frequency trader’s apple cart – yes? I must admit I am not a stock market guru, but I can’t figure out exactly how a scheme like you are describing could be successful for very long if the other side of the high frequency trading knows about it.
LikeLike
Mike Kimel at AngryBear has done a lot of econometric analysis that shows that US economic growth has been highest when the top federal income tax marginal rate has been roughly 60-70 percent. He hypothesizes that when rates are high, businesses put more money into making themselves bigger and better (and more valuable), extracting less money for the owners’ other purposes. Anecdotally, businessmen seem to think this is an obvious truth.
LikeLike
Hi Frank, thanks for the detailed reply.
I’d have to think about your tax idea more before I could intelligently comment on it.
As far as the HFT point, the HFTs don’t hang onto securities for very long, maybe a few minutes, so the market risk is small (they shut off if it gets too volatile, like in the flash crash).
But the HFTs have huge built in advantages in steady markets. They can see orders before anyone else, either due to “flash trading” or due to fast computer servers located next to the exchange. They make their 2bp (on average, not every time, and the amount is a conjecture) by adjusting their trades before anyone else even *sees* the new orders. That’s why they spend money on a computer arms race; they want to be faster than even the next fastest HFT. They benefit from a market structure that rewards millisecond speed. This crowds out slower market makers, which means markets go haywire if the flash machines ever turn off.
I hope that makes sense; it’s arcane stuff. I focused on HFT so we may have been arguing over different things. I don’t really have time to continue on this thread.
PS
I think it would be worth studying the history of this stuff. I suspect people used to employ messengers (horse first, rail and telegraph later) to transmit information (such as ships arriving in port) to stock exchanges. So the whole “financialization” hullabaloo probably isn’t as new as it seems. There were always expensive resources devoted to competitive trading.
LikeLike
One of the most important and best posts I’ve ever read. Of all the many many posts I’ve read over the years this is the first one to note Hirshliefer’s paper,and I can’t right now recall a post really directly noting the concept expressed in that paper, although I’ve talked about it in comments, the zero-sum, gambling nature of so much of the great wealth finance people obtain.
It’s amazing and tragic that a big name like Hirshliefer formalized this 41 years ago and it’s so little known today.
LikeLike
Steve,
I would really like to here your thoughts on the sale of tax breaks.
Everyone knows that discretionary policies (taxes, spending, interest rates, etc.) are typically bad economics because people in general like a fair amount of certainty in their lives.
From Friedman’s permanent income hypothesis to Larry Summers recent article:
http://www.huffingtonpost.com/2012/03/26/larry-summers-recovery_n_1379203.html
“Contingent commitments have the virtue of providing clarity to households and businesses as to how policy will play out, and in areas where legislation is necessary, eliminating political uncertainty. They allow policymakers to project a simultaneous commitment to near-term expansion and medium-term prudence – exactly what we require right now.”
In essence Larry Summers is talking about Paul Krugman’s “confidence fairy”.
Tax policy has never been structured to take this into account.
Instead tax policy has been all over the map. Part of this comes from a social view of tax policy (tax breaks for the rich), part of it from a political view (Wanninski’s two Santa Claus theory), but little of it from a pure economics point of view.
From a pure economics point of view, tax policy should compliment monetary policy with both geared toward reaching goals. And the only way you get a commitment from an enterprise that is subject to political change and upheaval is through contracts (something Newt Gingrich capitalized on with his “Contract with America”).
When U. S. Treasury sell tax receipts, it has made a contingent commitment to the owner that at maturity the receipt will be accepted for the payment of taxes.
I realize that this goes well beyond what we were previously discussing (high frequency trading), but I also think this is a worthwhile subject to be discussed.
LikeLike
Bill, I was not suggesting discouraging arbitrage, I am suggesting discouraging an (approximately) socially useless investment in information discovery that results in large transfers of wealth by the creation of information advantages that would not exists but for the investment in information. I think, as some commenters have pointed out, that there are very large empirical questions about the magnitude of the investment in such “research” activity and how much of the growth of the financial sector is traceable to such activity. But some such activity is taking place and it’s a pure social waste. But it goes way beyond “arbitrage.” I am afraid that I don’t see the connection between this and your discussion of the payments system. Care to spell it out for me?
Enourmousturnip, My argument was not related to financial stability issues at all. It was purely a question of whether the large profits accruing to the financial sector over the past 30 years correspond to socially productive activities on their part or whether the gains of the financial sector are being derived by diverting resources from resources that really are socially productive. There is no cyclical implication that I deduce from this argument.
Benjamin, Back in those days, people often did not distinguish between average and marginal tax rates. Supply-siders actually did force everyone to address more clearly the distinction between average and marginal rates. Clearly someone who “got” $1million would not have owed $900k in taxes.
Frank, The Laffer curve is a simple theoretical relationship between tax rates and revenues. The only policy implication is that for any level of revenue that could be collected, you should collect it with the lower of the two rates that would generate that revenue because taxes discourage the production of socially useful output. One way to read my post is as a challenge to the assumption that all private economic activity is socially productive.
Steve, I agree that I was oversimplifying. I was just trying to explain a concept. I tried to make it clear that I am not at all sure that the theoretical issue that I am drawing attention to is empirically important in accounting for the growth of the financial sector as a whole. I don’t know enough about high frequency trading to have an opinion about whether it is socially productive or not.
You are totally right that the concept that I was discussing is applicable to a much wider range of activities than finance. Note that the words “inventive activity” were included in the title of Hirshleifer’s paper. I almost included a discussion of the increasing importance of “intellectual property” over the past 30 years in my post, but that would have expanded the post a lot and taken more time and energy than I wanted to expend. But I think that much if not all of what I said about finance applies to IP as well.
Bob, Thanks for providing that very relevant example of what I am talking about.
Floccina, Yes it might. I haven’t thought about how to go about estimating how much different incentives to engage in financial transactions have accounted for the observed growth in the financial sector.
Sir Algernon, We know almost nothing for sure. I am just drawing attention to a possibility and urging people to think hard about whether it is empirically relevant.
Do you know for sure how many useful financial products have been created in the last 30 years?
The fact that people are willing to trade on these exchanges does not prove that the existence of an entire industry that supports such trading is socially useful. That’s the whole point of Hirshleifer’s paper, a misalignment between private and social incentives. And as I pointed out above to Steve, the argument applies to IP activity as well, an area of the law that has become totally dysfunctional.
Tas, The usual presumption is that people become rich in a capitalist society by producing value for the rest of society. But what if there is a large group of people who are becoming rich by destroying value for the rest of society?
PJR, I have trouble imagining that you can derive statistically meaningful results about the effect of marginal tax rates from the available data. If we had 500 years of data, maybe I would begin to take such results a little bit seriously.
Richard, Thanks for your kind words. Actually I am a bit surprised to hear how surprised people have been by the argument of this post.
LikeLike
Consider poker.
Five people play once a week. They are all very productive and big gamblers. One of them wins regularly and the other four lose. The winner quits his job and just lives off of the poker winnings.
Total output of our 5 person economy drops by about 20%. Total income drops by 20%. The “winner” earns a personal income equal to the others, but is supported by them. (At least, that is how I think it how gambling winnings and losses are measured.)
Now, suppose the 5 are enemies and are constantly arguing with one another about the boundaries between their property. Two of them are really good at making these arguments. The other three begin to pay the other two to argue for them. They are so sucessful and arguing, the quit work. Total output for the 5 person economy drops 40%.
Oh, but no. The two arguers are “lawyers,” and their output is legal services. Total output is unchanged. The lawyers time is spent producing legal services. The other three produce what they were before. They are still supporting the two lawyers. At least, that is how I think it is measured.
Now, rather than poker, suppose that they play, “guess the temperature.” It is kind of complicated. A guess is made for tomorrows temperature and someone who thinks its too low will bet that it will be higher. But nothing happens unless someone else takes the bet, betting that it is too low. If there is no action, another guess is made. It is so complex, that one of the players quits his job and just spends full time being the bookie. He takes the bets and pays everyone off.
Like with poker, you might think that output dropped 20%. But no, the bookie is providing finanical services for the futures market. Total output is the same.
Economists are notorius for proposing that investors diversify and buy and hold. We say, don’t try to pick stocks. Don’t try to time the market.
But many people do. Is it like poker? Is it like my futurers market?
In the first story, anyone who gets tired of supporting the “winner” in the poker game can quit playing. Similarly, anyone who doesn’t want to play “guess the temperature” can quit.
With the lawyers, If they really determine the division of property based upon the strenghth of their arguments, no indvidual can avoid playing. Having clearer rules to start with, so there is less to argue about, would be the only answer.
Now, do people who diversify and buy and hold suffer losses due to the “finanical casino?” Or is it those who refuse to listen to that counsel and instead try to “beat the market” by “buying low and selling high,” start playing the game. Are the high income earners like some combinintion of the person organizing the temperature bet and the poker player?
Or is it more like the lawyers? It is some problem with the rules that shifts resources to rent seeking (including defense?)
LikeLike
David,
“The Laffer curve is a simple theoretical relationship between tax rates and revenues. The only policy implication is that for any level of revenue that could be collected, you should collect it with the lower of the two rates that would generate that revenue because taxes discourage the production of socially useful output.”
I kindly disagree. I don’t believe that the Laffer curve has any policy implication simply because tax rates are a fixed cost rather than a floating cost. As such, they have no predictive power over the type of output that is produced.
I said as much above.
“One way to read my post is as a challenge to the assumption that all private economic activity is socially productive.”
I am not sure what you mean by socially productive. Would you care to elaborate? How is social productivity measured?
LikeLike
it’s unclear to me why lower tax rates on the wealthy particularly blow up the finance sector, as opposed to other sectors in which the participants are generally high-income.
LikeLike
Here’s why lower tax rates blow up the finance sector.
Define “wealthy” as those who have income beyond their needs/desires for nonfinancial goods and services so that 100% of any increase in their wealth/income will be available for investment.
In the final analysis the true value of investment goods must be directly related to the income that can ultimately be generated from them and so will be proportional to total output (GDP). If taxes on the wealthy are cut they should be able to accumulate wealth at a faster rate than previously and a rise in the fraction of wealth held by the wealthy should rise. This has indeed happened over the last 30 years.
In particular, the ratio of wealth held by the wealthy to GDP should rise, meaning that the money value of assets should rise relative to their true value (i.e. the amount of real output/income that can be generated from them). If this is true we should see evidence of rising valuations of assets such as secular falling dividend yields on the S&P500, more frequent and larger financial bubbles and associated financial crises and asset value crashes. These things have also happened over the past 30 years.
A period of rising asset prices is favorable to speculative activities. Rising asset prices should also stimulate supply. Thus the last 30 years should have seen a secular increase in the faction of the workforce engaged in financial activities and should also have seen the creation of new types of investment products to tap rising demand for assets created by increasing pools of investable cash and rising “animal spirits” caused by an ever-lengthening period of price appreciation. These things have also happened.
Finally there was deliberate government policy in the 1980’s that created a world in which the real return from government bonds exceeded 5%, the long-term real return on equity in the US. That is, for an extended period of time, it was more profitable for an individual to sit on his butt and clip coupons from risk-free bonds than to engage in actual enterprise. I can think of nothing more effective at goosing the growth of the financial sector than this set of affairs.
And once goosed, the facotrs listed above would reinforce the trend.
It has come to pass that (mostly tax) policy over the last 30 years has restarted the financial panic cycle that had lain dormant for three quarters of a century after 1933. We now face the likelihood of recurrent 2008-type events with a 14 +/- 4 years periodicity going forward, assuming the old timing is still valid, which I see no reason to doubt since the proximal cause of panics is probably psychological in nature and hence unchanged from the pre-1933 world.
LikeLike
This is extremely silly. High marginal rates will at a certain point increase tax avoidance, because costly tactics (lawyers, time, legal risks) become more valuable at the percentage go up. The logic here is that low marginal rates do the opposite, by making “information hunting” more profitable the total costs aren’t high enough to prevent it.
But as everyone recognizes, information hunting (Finance) is EXTREMELY valuable. No matter how high the tax price is, it’s going to still be profitable. Raising marginal rates will eliminate activities that are LOW margin (difference between profit and cost is close to zero), because those are the things that are close to being net negative.
Capital gains differential likely has a much bigger effect on our shift from investing in human capital (doctors, lawyers, etc.) towards investing in knowing how to allocate physical capital (finance).
-AFG
afoolsgame.com
LikeLike
Bill, you are highlighting very effectively the inconsistencies and paradoxes associated with measuring national income. Much of what is spent or earned represents no net social product. Was US military spending aimed at defending Europe from Soviet military expansion a contribution to US GDP? The answer depends on an evaluation of what the consequences of a US withdrawal from its defense commitment to Europe would have been, and how those consequences would have affected the welfare of Americans. No one even thinks about trying to answer those questions. Instead military spending is just included in GDP and that’s the end of it. Similarly, lawyers services are treated as a net addition to GDP even though litigation is a zero-sum game. Nevertheless, I have no doubt that from the point of view of society there is a huge net loss by employing a physicist in an investment bank or a hedge fund to devise trading strategies instead of in an inner city high school teaching physics.
Frank, By socially productive, I mean activity that adds value to one person without destroying value to another. If I teach a trader how to make a larger profit the value he gets from me is offset by the value lost by the traders from whom he extracts his profit. There is no net creation of value. If I produce a final good or service that is valued for its own sake, I have added to the total value created by society.
irreparailetempus, I did not fully explain that point. It is just a conjecture. Actually, “blow up” is your term. I simply referred to expansion. I see that in some discussions on the web today, people are interpreting me as attributing the financial crisis to reduced marginal tax rates. That is not correct. I was talking about a possible long-run tendency not a cyclical effect. The question is not whether there are people earning high income working in a particular sector. The issue is whether the high incomes being earned are related to socially productive or socially unproductive activities. I assert that all forms of research aimed at trading profits and other activities geared to improving trading strategies are socially unproductive because they generates profits for one side of the transaction and offsetting losses for the other side. I think that there are a lot of high income earning activities that are associated with zero sum unproductive activities. What I am also asserting is that the high incomes earned in these sectors is inducing talented workers from other sectors to seek employment in finance with a corresponding net loss to society.
Mike, You are implicitly assuming either that asset markets are incorrectly valuing assets or that discount rates are not falling to reflect the increase in net saving as a result of increasing income inequality. You could be right, but you have to specify why markets are systematically mispricing assets.
AFG, Well it may be extremely silly, but there is also the possibility that it is not silly, but that you have not understood my point (actually Hirshleiffer’s who published it over 40 years ago) that there is a systematic difference between the private and social value of information that can be used by traders. Information is much more valuable to an individual who has information that others don’t than the value to society if the information were freely available. That difference creates an incentive to gather information that is only valuable as long as the information is kept private.
LikeLike
David,
That seems like a very accurate description (seriously). But what does it have to do with marginal tax rates on labor income?
“So I am inclined to conjecture that over the last 30 years, reductions in top marginal tax rates may have provided a huge incentive to expand the financial services industry.”
That’s the silly part.
LikeLike
David,
The stock market at least clearly misprices equity:
http://www.safehaven.com/article/69/relative-pr-a-novel-valuation-tool
The reason why is simple. Stock values are based on future performance, which is unknowable, thus it is impossible to assign a correct value.
What is done in practice is to use some sort of model or paradigm to provide a value. These paradigms tend to be overly optimistic or pessimistic at times, usually depending on how the present economy is performing. During periods of economic turmoil, such as now, the 1970’s the 1930’s and so on, pessimism rules and valuation gets increasing pessimistic. At other times such as the 1990’s 1960’s, 1920’s etc. the opposite happens. The result is alternating secular bull and bear markets.
http://www.safehaven.com/article/68/secular-market-trends
I developed P/R, a valuation tool for identifying these secular trends, and used it to make a forecast in 2000 that stocks would not beat money markets for the next 20 years:
Since then I’ve had some success with identifying market bottoms (not tops):
In 2002: http://www.safehaven.com/article/84/how-low-can-we-go-what-several-valuation-methods-have-to-say
In 2008: http://www.safehaven.com/article/11887/stock-cycles
Taken together there is solid reason to believe that the market mind gets it wrong in a roughly predictable fashion.
The cause of this, I believe is a lagged negative feedback loop between observations of events leading to paradigm creation and use of paradigms to formulate policy which affects events.
http://www.safehaven.com/article/2975/the-paradigm-cycle-model
Thus events influence events, sort of like temperature influences temperature in your home heating system that uses a negative feedback loop to control home temperature.
Suppose you introduce a lag between when the furnace responds to the temperature signal. The result would be the furnace would stay on after the high temperature was reached heating the house further. After the lag it would then shut off and the house would cool. When the temperature reached the lower set point the signal would be send to the furnace to turn on, but the lag would delay the response so the house cools further before the furnace finally turns on. The result is cycling temperature.
This result is general, any lagged negative feedback loop will produce cycles. In the economy, policy affects events. Policy is strongly influenced by the paradigm/model used to systemize economic thinking. The basic framework for a worldview or paradigm is picked up in rising adulthood as one first develops an understanding of how the world works.
Thus you have events (temperature) affecting paradigm development (thermostat) in one’s 20’s.
Policy is determined by people in late middle age, so 30 years after paradigm acquisition (the temperature signal is received by the thermostat) the paradigm is used by people in their 50’s to affect policy (furnace) which affects events (temperature). The result is the same as the lagged thermostat: cycles. This gives rise to periodic economic/political/social crises such as now, and in the 1970’s and in the 1930’s, and the 1890’s and the 1860’s.
LikeLike
Mike Alexander,
“Finally there was deliberate government policy in the 1980′s that created a world in which the real return from government bonds exceeded 5%, the long-term real return on equity in the US. That is, for an extended period of time, it was more profitable for an individual to sit on his butt and clip coupons from risk-free bonds than to engage in actual enterprise.”
All that I can say is that it takes two to tango. For someone to buy government bonds that deliver guaranteed 5% real returns, the federal government must sell those bonds. Yes, the federal reserve and the “bond market vigilantees” pushed real returns on government bonds high during the early 1980’s. And what also happened during the 1980’s is that the federal government ran persistantly large budget deficits to facilitate the sale of those bonds.
The question you must ask yourself is should the federal government sell guaranteed liabilities at all. Most of corporate America uses a mix of equity and debt in financing their deficits. Should not the federal government do the same?
LikeLike
Frank wrote: For someone to buy government bonds that deliver guaranteed 5% real returns, the federal government must sell those bonds. Yes, the federal reserve and the “bond market vigilantes” pushed real returns on government bonds high during the early 1980′s. And what also happened during the 1980′s is that the federal government ran persistantly large budget deficits to facilitate the sale of those bonds.
A government overseen by a self-described fiscal conservative and advocate of “small government” slashed taxes while expanding government cost, chiefly in the military sphere, resulting in very large deficits during the 1980’s expansion. This is precisely the policy to which I referred.
At the time Reaganomics represented a political stategy innovation for Republican party as previously fiscal-conservative Republicans so feared the specter of inflation that they supported higher taxes if necessary to prevent inflationary deficits.
With the policy innovation of crushingly high interest rates, it was possible to have low inflation with low tax rates–all without cutting spending with its associated political pain. The outcome was (1) more rapid deindustrialization (2) asset inflation leading to an increasing size of the financial sector and (3) the eventual return of the panic cycle. What made the policy brilliant was that the bulk of the costs fell on Democratic constituencies while the benefits flowed chiefly to Republican constituencies.
It was the development of this brilliant political strategy that led to three decades of Republican dominance in policy debate and made Reagan a deified figure in Republican history. Unfortunately, side effect #3, which could not be foreseen (at least I did not do so) threatens to put a serious kink in Republican political fortunes.
Democrats went through the same thing with their version of Reaganomics, insitituted during WW II by FDR. Their policy promoted rapid industrializaiton and strong wage growth across all income quintitles, through a high-tax, low real interest rate, balanced budget policy. This came at the expense of the investor class, who faced an unappetizing combination of high taxes and low real interest rates that made it very difficult to grow wealth through passive investing. Here it was Democratic constituencies, the working class and unions concentrated in the industrial sector that benefited and rentiers (a Republican constituency) that suffered. And FDR gained the same sort of deification that Reagan has in his party as a result of the political success of his innovations.
Both economic packages are subsets of larger paradigms that arose in response to the crises of the previous decade. Both were effective for about three decades becoming less and less effective as economic aspects each policy neglected grew in importance, eventually leading to failure, creating another crisis, which will bring about a new paradigm that will address the problems of today, and then go on to fail some four decades down the road.
LikeLike
“Democrats went through the same thing with their version of Reaganomics, insitituted during WW II by FDR. Their policy promoted rapid industrializaiton and strong wage growth across all income quintitles, through a high-tax, low real interest rate, balanced budget policy.”
I disagree with your statement on real interest rates. Here is a plot of real AAA 30 year rates over the last 90+ years:
http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=CPIAUCNS_AAA&transformation=pc1_lin&scale=Left&range=Custom&cosd=1920-01-01&coed=2012-02-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2012-04-05_2012-04-05&revision_date=2012-04-05_2012-04-05&mma=0&nd=_&ost=&oet=&fml=b-a&fq=Monthly&fam=avg&fgst=lin
I am using the 30 year AAA because it is one of the few interest rates that the St. Louis federal reserve has data for back to the 1920’s.
There are two significant periods where real long term interest rates were significantly negative and stayed negative over a protracted period:
1941 – 1952
1973 – 1981
That is about 21 years out of about 90 starting in 1920. In fact the average long term real interest rate is about 2.8% including the Great Depression of the 1930’s and the Great Inflations of the 1940’s and 1970’s.
I don’t believe you can point to a Congressional Democrat versus Republican view of interest rates. Instead you need to look at who was running the federal reserve at the time and his political / economic views.
1941 – 1952 was primarily a mix of Mariner Eccles and Thomas McCabe, William Martin became president in 1951
1973 – 1981 was primarily Arthur F. Burns, with a brief stint by G. William Miller, Paul Volcker became president in 1979
LikeLike
Frank,
If you look at your graph you can clearly see my point. If you look at the 1983-2000 period, the time when the Reagan paradigm was at its most effective, real intest rates were in the 4 to 8 range.
Now look at the 1950-1970 period, the heyday of the FDR paradigm. Real interest rates ran at 0 to 4 range.
In this comparison I am presenting the periods when each paradigm was at its height of influence and effectiveness. So its apples and apples.
My more comprehensive analysis is here:
http://www.safehaven.com/article/11230/stock-cycles
Paradigms are NOT ideologies. The Reagan paradigm has both left and right aspects, as did the FDR paradigm. For example, gay marriage is part of Reagan paradigm, because the same social forces that made Reaganomics possible also enabled the gay liberation movement. Similarly, the military-industrial complex and national security state were parts of the FDR paradigm, even though support for these are now a top priority of the Right.
When a crisis hits, some elements of the reigning paradigm are overturned, while others are preserved. So whereas Keynesian economics, high tax rates and peacetime conscription (all features of the FDR paradigm) were discarded, the military-industrial complex and welfare state were retained under the Reagan paradigm.
Paradigms are ways of thinking about how the world is (or should be)organized. The Reagan paradigm and the paradigm before the New Deal were both examples of freedom-type paradigms. The New Deal and post-Civil War paradigm were progress paradigms.
From my paradigm article:
The origins of the paradigm mechanism for cycle generation were in the aftermath of the Revolutionary War secular crisis, in the emergence of the first political divisions in the new nation. The first “Progress” paradigm arose out of support for George Washington, hero of the Revolution, and was based on the progressive ideas of Alexander Hamilton, Washington’s aide during the war and Secretary of Treasury in his administration. The first “Freedom” paradigm was based on the libertarianism touted by Thomas Jefferson in opposition to Hamilton’s Federalists.
Now that a crisis has arisen the two parties are each touting modifications of the existing Reagan paradigm as potential solutions. If the range of possible modifications are large enough it is possible that an effective soltuion could emerged from an appropriate re-arrangment. I do not think the range of economic policies is near large enough and so no solution is possible at this time. New memes need to be thrown out there and catch fire. One such meme is the 99% vs the 1%. It brings back elements of class struggle to the political discourse that have been absent for a long time and widens the range of permitted thinking.
The idea that higher taxes might actually be good for growth (something I suspect is true) would be an excellent meme to be “out there” and so I was happy to see this blog post–although its years past schedule. I would also like to see ideas like an across the board tariff to get out there. Free trade has been part of the last two paradigms and so it pretty stale. I suspect its gonna have to go if we are to get an effective new paradigm. I have a hunch that the empire is going to have to go too, so I was heartened by the success of the Ron Paul campaign at raising some of these issues amongst the younger generation (who will adopt whatever new paradigm, if any, that emerges out of this crisis era.
An effective new paraidgm does not have to rise. It did not for imperial Spain after the late 17th century crisis period, and a century of precipitious national decline was the outcome. Somehting like this could very well be in store for us if a sufficiently broad range of memes are not activated.
LikeLike
Mike,
I appreciate your reflections on economic paradigms. I too was cheered by Ron Paul’s campaign success in rejecting the U. S. as an imperialist state.
I disagree that a new paradigm will arise out of higher tariffs and taxes.
Congress actually acknowledged the short comings of the “FDR paradigm” in 1978 through the Humphrey Hawkins act. And so if you are looking for a new paradigm, you might start there.
I would predict that a new economic paradigm will arise from a capitalist view of government. Mr. Reagan was correct that deficits don’t matter – never have, never will. The sole reason that the capital markets (stocks and bonds) exist is to permit the financing of deficits.
Where Reagan went off course was in not realizing that the federal government need not sell debt to finance its deficits any more than IBM or General Electric or JP Morgan has to sell debt to finance its own.
Most major corporations have equity financing available to them, and it stands to reason that the federal government should be able to sell equity like claims on its own revenue stream (taxes).
In the corporate world equity and debt have some distinguishing features:
Equity is a claim on profitability, debt is a claim on cash flow / net worth
Equity is a time variable claim, debt is a fixed duration claim
Equity is a junior claim, debt is a senior claim
I am not suggesting that the federal government be run as a for profit enterprise. What I am suggesting is that the federal government need not sell all guaranteed (senior) claims on its revenue stream (taxes). It could just as easily sell non-guaranteed (junior) claims – equity like claims,
LikeLike
I don’t see the difference between senior and junior claims for a government which cannot go bankrupt. The claims you talk about seem to me toi be the same sort of thing as the British consols or 18th century government stock. It’s the most expensive form of debt; I see no advantage to the issuer.
As to the composition of the next paradigm this is unknowable as all of use are at present blinded by the current paradigm and so cannot see it. My point about higher taxes and tariffs iosn’t that there would be good policies, but that they increase the range of ppolicy options and will create a motivation for opponents of these things to solving the problem (instead of ignoring it for 30 years as they have done).
As for the 1978 Humphrey-Hawkins Act, of course it was acknowleded that the New Deal paradigm had shortcomings. This was the reigning paradigm in the midst of the crisis that invalidated it. Humphrey-Hawkins was a step towards the new paradigm that replaced it. Now this paradigm is failing, and what is needed is for a Republican house to acknowledge that shortcomings of Reaganomics. Hasn’t happened yet so we have a long way to go.
LikeLike
Mike: “I don’t see the difference between senior and junior claims for a government which cannot go bankrupt.”
The difference does not come in the affect that it has on the government’s
finances. You are correct, the federal government cannot go bankrupt. The difference comes in how those junior claims affect private sector finances.
The bond market is structured toward credit risk, federal government bonds are AAA and as such, the federal government’s cost of debt service is lower than the private sector. That would not be a problem if we weren’t constantly waging war and putting those wars on the national credit card.
In addition, on a pretax level, the cost of debt service in the private sector can never be below zero. This has significance when you look at the Great Depression from a debt deflation lens.
During the Great Depression, interest rates were relatively mild on a nominal basis (around 2%), but severe on a real basis (upwards of 10%).
Where Reaganomics screwed up was in not fully understanding how tax policy can be adjusted to allow for high real interest rates and deflation without the corresponding unemployment and poverty.
On an after tax basis, the cost of debt service in the private sector can be negative, in fact severely negative.
Instead, in 1983 the bureau of labor statistics eliminated housing prices from the Consumer Price Index and replaced it with owner’s equivalent rent. This policy lapse led to a mini-housing bubble in the 1980’s, followed by a major housing bubble in the 2000’s.
As for the Humphrey Hawkins Act, it had four goals:
No Trade Deficit
No Federal Budget Deficit
Full Employment
Low Inflation
Since its passing there have been 5 presidents – Reagan, Bush I, Clinton, Bush II, and Obama
Reagan hit on full employment and low inflation but missed on twin deficits
Bush I hit on twin deficits but missed on inflation and employment
Clinton hit on employment, inflation, and budget but missed on trade
Bush II hit on employment but missed on inflation, budget, and trade
And Obama has yet to hit on anything
What all five missed on was that deficits need not be financed with debt.
LikeLike
Mike,
The British Consols that you spoke of were infinite duration bonds (aka perpetual bonds). They paid an interest rate indefinitely. As such they were equity like because they did not have a fixed duration. British Consols were not junior to fixed duration British bonds. Interest payments on both kinds of bonds were treated the same.
I am referring to a different characteristic of equity – equity is a subordinate (non-guaranteed) claim on a corporation’s revenue stream. They are subordinate because they are claims only on profits, not total cash flow. Corporate bonds are a higher level claim because untimely or nonpayment of debt forces a company into some form of bankruptcy and sometimes liquidation.
LikeLike
I still don’t wee why subordinate vs not subordinate makes any difference for an entity that cannot become bankrupt.
I’ll try with an example to see if I understand where you are coming from. Suppose the Federal government were to levy a new 4 cent/gallon tax on gasoline and distillate fuel oil consumed in the US. Such a tax would bring in about $2.8 billion today compared to about $3.4 billion before the recession. Suppose the government offered to sell this income stream to private investors for, say $60 billion, as an equity-like asset. The government would then use the $60 billion to retire currently expiring long-term bonds.
if the government cannot get the $60 billion it simply uses the proceeds of the tax to pay down debt at a slower rate and tries the capital markets at a later date (say after the economy recovers a bit and the income stream has shown strong growth, which should boost its multiple.
Since 4 cents is only 1% of the current price, natural inflation should be several times this and the government ought to be able to do this every year to retire another $60 billion worth of expiring debt and so whittle the debt down a little every year. The government could (and does) retire the old debt by borrowing the $60 billion each year (at the current very low rate), but then the old debt would be replaced by new debt and the deficit would not change.
Is this the sort of thing you are talking about?
LikeLike
Mike,
No, this is not the sort of thing I am talking about.
“The difference comes in how those junior claims affect private sector finances.”
Suppose the federal government comes along and sells you a receipt for taxes that you pay now that are due 30 years from now. Suppose to entice you to buy that receipt the federal government tells you that the receipt will return 10% annualized over the next 30 years. That 10% return is not guaranteed. The only way to realize that 10% annualized return is to have a tax liability equal to or greater the value of the receipt 30 years from now.
And so the federal government has just sold you a subordinate claim on its revenue. While a 30 year bondholder will always get his / her money back with interest, you might not if you have no tax liability 30 years from now.
Now suppose at the same time you have a 30 year debt (mortgage, company, whatever) that you are paying 5% on. If the initial value of your debt (how much you borrowed) is equal to the initial amount of money that you spent buying the receipt, then on an after tax basis, your cost of debt is around -5% (actually less than that if your debt is amortized).
“During the Great Depression, interest rates were relatively mild on a nominal basis (around 2%), but severe on a real basis (upwards of 10%).”
If the after tax cost of debt in the private sector can be less than 0%, then why is inflation required?
LikeLike
“Since 4 cents is only 1% of the current price, natural inflation should be several times this and the government ought to be able to do this every year to retire another $60 billion worth of expiring debt and so whittle the debt down a little every year.”
There is nothing natural about inflation. It is the effect of poor economic policy.
LikeLike
Frank you did not aswer the question. Was the example I gave what you were talking about. If not, please provide one.
LikeLike
Mike,
Read the first reply, starting with:
“No, this is not the sort of thing I am talking about.”
The example that I give in that reply describes a subordinate claim on tax revenue.
LikeLike
Frank,
I scrolled down to the bottom and did not see your first reply. Sorry.
Inflation is still built into your concept. What you are doing is asking for future taxes to be prepaided (at a steeply discounted value). We can imagine all sorts of securities of this sort being issued with a variety of durations.
In the short run the issuer would raise additional income. But as soon as the first of these securities come due the issuer would suffer a drop in income far larger than the income boost orignally obtained. This creates a much larger deficit problem than the original deficit problem for which the security was issued. Dealing with deficit using debt or monetaization would lead to inflation.
Trying to use more of the subordinate securities won’t work because most of the future tax liability in future years would have already been sold. The only way out that avoids inflation would be to cut spending.
If cutting spending is an option, why not simply administer the spending cut avoided by issuing the security in the first place and so avoid the much larget future cut?
I still don’t see any utility to idea.
LikeLike
“Inflation is still built into your concept. What you are doing is asking for future taxes to be prepaided (at a steeply discounted value). We can imagine all sorts of securities of this sort being issued with a variety of durations.”
No inflation is not built into the concept. You can have one of two views of inflation:
1. Inflation is a rise in the price level.
2. Inflation is a rise is the amount of money. And since all money begins as a debt – inflation is a rise a rise in the debt level.
By selling tax breaks, the federal government is able to reduce the total amount of debt outstanding. Imagine if the federal government sold more than enough receipts to cover current year spending, and used the excess funds to buy back its bonds from the federal reserve system. What would happen to the money supply? It would contract.
“In the short run the issuer would raise additional income. But as soon as the first of these securities come due the issuer would suffer a drop in income far larger than the income boost orignally obtained. Trying to use more of the subordinate securities won’t work because most of the future tax liability in future years would have already been sold.”
This is simple math. The present value of all future tax liability on an infinite time line is infinite. The reason we use an infinite time line is that taxes are a legal obligation, not the subject of choice. The federal government is not some company trying to sell hoola hoops. Its revenue stream is not predicated upon consumer whims that change over time.
As soon as the first of these securities comes due, the federal government would sell more of these securities. The federal government already does this with bond issuance, and so this is not a stretch. The reason the federal government can sell more of these securities out to infinity is because the present value of all taxes collected is infinity.
“The only way out that avoids inflation would be to cut spending.”
The only way that avoids inflation is to just not sell any debt, with indifference to spending.
“If cutting spending is an option, why not simply administer the spending cut avoided by issuing the security in the first place and so avoid the much larget future cut?”
It seems you are trying to argue from both sides:
“Trying to use more of the subordinate securities won’t work because most of the future tax liability in future years would have already been sold.”
If you believe that there is a limit to the amount of subordinate claims the federal government can sell, then you must believe there is a limit to the number of senior claims (bonds), that the federal government can sell.
“I don’t see the difference between senior and junior claims for a government which cannot go bankrupt.”
But then you say that the government cannot go bankrupt, which means that there is no limit the amount of bonds the federal government can sell.
Which do you really believe?
“I still don’t see any utility to idea.”
The utility to the idea comes from how this affects the after tax cost of money in the private sector. From above:
“your cost of debt is around -5% (actually less than that if your debt is amortized).”
If you agree that the federal government cannot go bankrupt (for the reasons above), then why would you object to the federal government selling a tax break to you that lowers your after tax cost of debt?
LikeLike
Frank wrote:
This is simple math. The present value of all future tax liability on an infinite time line is infinite.
This is only true in a special case. The series of interest can be represented by equation 1:
1. PV = P ∑ (1+g)^i/(1+r)]i for i = 1 to ∞
Here PV is the present value of future tax payments which in the current year are equal to P and then are assumed to grow at rate g afterward asa result fo economic growth. The discount rate is r.
This is convergent series with value equal to
2. PV = P (1+g)/(r-g)
To see this make the substitution z = (1+g)/(1+r)
Then equation 1 becomes
3. PV = P { ∑ z^i for i = 1 to ∞ }
You can look up the value for the series in brackets here:
http://en.wikipedia.org/wiki/List_of_mathematical_series
It is give as z / (1-z). Substitute (1+g)/(1+r) for z and it reduces to (1+g)/(r-g)
For the special case where r = g the series is infinite. Any r value attractive to investors will be greater than g, for which a finite vlaue can be obtained.
For example you suggested 10% for r. GDP growth over the long run is about 3% so we use this for g. Plugging these values in equation 2 gives:
PV =P (1+0.03)/(0.10-0.03) = 14.7 P
That is, the present value of all future tax payments with a 10% discount rate is enough to run the government for around 15 years.
LikeLike
I made a typo, equation 1 should be:
1. PV = P ∑ (1+g)^i/(1+r)^i for i = 1 to ∞
LikeLike
Mike,
A couple things. First, you really can’t use a non-guaranteed rate of return as a discounting method. Notice that I didn’t say interest payments on government debt (guaranteed by the 14th amendment) should be 10% annualized. What should be obvious is that some people who buy forward year tax receipts may not realize all of the gains at the end of the holding period.
Second, the 10% that I mentioned was at the long end of issuance (out thirty years). As you mentioned – “We can imagine all sorts of securities of this sort being issued with a variety of durations.”, there would be a variety of maturities. And like Treasury bonds, shorter duration instruments would have lower potential returns. And so, even if you wanted to use a non-guaranteed rate of return as a discounting method, you would need an average across the maturity spectrum. Assuming an even distribution of maturity issuance, this would be an average potential rate of return of 5%, with 10% at 30 years.
Third, federal government liabilities (bonds currently) are accrual type rather than constant payment (coupon) type. As such, the principal and interest are returned to the investor at maturity. Currently, the longest dated bond the Treasury department sells is 30 years, but 30 years is really an arbitrary number. The federal government could just as easily sell 50 year or possibly 100 year securities.
Fourth, Real GDP growth has average about 3% over the last 80 years in the United States.
http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=GDPCA&transformation=pc1&scale=Left&range=Custom&cosd=1929-01-01&coed=2011-01-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2012-04-16&revision_date=2012-04-16&mma=0&nd=&ost=&oet=&fml=a&fq=Annual&fam=avg&fgst=lin
Nominal GDP growth has average about 5.5-6% over the last 80 years in the United States.
http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=GDP&transformation=pc1&scale=Left&range=Custom&cosd=1947-01-01&coed=2011-10-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2012-04-16&revision_date=2012-04-16&mma=0&nd=&ost=&oet=&fml=a&fq=Quarterly&fam=avg&fgst=lin
I imagine that you were referring to real GDP at 3% growth.
Fifth, if you plot real gross domestic product and real interest rates (pick your duration), you will notice a strong correlation (markedly so after the 1970’s). There is causation built into the correlation. The reason for this is the way the banking system changed in the late 1960’s and through the 1970’s. The primary movement was the securitization of debt. Rather than banks holding onto all of the loans they make, they began selling them. And so, interest income became more widely shared through an economy (CD’s, mutual funds, etc.).
Sixth, if you hold that as real interest rates climb, so will real growth, then the only thing missing is to make high real interest rates affordable. In 1983, the bureau of labor statistics did this on a limited basis – they eliminated housing prices from their measure of inflation. On an economy wide scale, to make high real interest rates affordable, a tax policy instrument is required.
And so I guess I have several problems with your calculation – it misses the positive correlation (and causation) between real interest rates and real growth, it misses that the 10% that I mentioned was neither guaranteed, nor was it an average rate of return across all maturities, and it misses that the federal government is not limited in the longest dated security it can sell. If you insist on using a non-guaranteed rate of return as a discounting method, then please include a fail rate – how many of the forward year tax receipts will not be fully realized by the owner at maturity, and use an average potential rate of return across a maturity spectrum.
Finally, I would say this. The past is a useful guide for economic performance, but is by no means a limitation. There is absolutely no economic reason the United States is limited to 3% real growth.
But then again, I am an optimist 🙂
LikeLike
Mike, reworking your equation:
PV = P x ∑ [ (1+g)^i/(1+r)^i ]
If we are going to use interest rates on debt as our discounting factor, then we need to factor in whether the overall government debt level is rising or falling. After all, it is easier paying 10% on 10 trillion dollars worth of debt than it is on 15 trillion dollars worth.
PV = P x ∑ [ ( 1 + g)^i / ( ( 1 + r ) x (1 + D) )^i ]
And so there are multiple solutions:
PV = P x [ ( 1 + g) / ( 1 + r ) x (1 + D) ] / [ 1 – ( 1 + g) / ( 1 + r ) x (1 + D) ]
PV = P x (1 + g) / (r + D + rD – g)
For the present value to equal infinity:
r + D + rD = g
For a growth rate of 3% and an interest rate of 10%, the federal debt would have to contract at a rate of:
D = (g – r) / (r + 1) = -6.7% annually.
LikeLike
Mike,
I have been running through your formula using a fail / success rate for tax breaks that are sold. Like I previously mentioned, because we are dealing with non-guaranteed claims on future tax revenue, we must accept that some owners will not get the full value of the asset at maturity – i.e. they may not have enough taxable income and tax liability to realize the full value of the asset.
Back to your original formula with a success rate added:
PV = P x ∑ [ ( 1 + g ) / ( ( 1 + r ) x ( 1 + s ) ) ] ^ i
r = rate of appreciation averaged across a maturity spectrum for tax breaks sold by the federal government
g = growth rate
s = success rate on tax breaks sold by the government
What I mean by a success rate is the amount of appreciation that the owners of said tax breaks are able to claim due to taxable income and therefor future tax burden.
We can say that the success rate is directly proportional to the rate of economic growth (g) and inversely proportional to the rate of appreciation (r).
And so s = g – r, plugging this back in for s gives us:
PV = P x ∑ [ ( 1 + g ) / ( ( 1 + r ) x ( 1 + g – r ) ) ] ^ i
Letting i go to infinity we get:
PV = P x [ ( 1 + g ) / ( ( 1 + r ) x ( 1 + g – r ) ) / [ 1 – ( 1 + g) / ( ( 1 + r ) x (1 + g – r ) ) ]
Rearranging we get
PV = P x [ ( 1 + g ) / ( rg – r^2 ) ]
To get the present value to equal to infinity all that we have to do is set
rg = r^2
r = g
Set the r for the rate of growth you hope to get, and sit back. Obviously, it is ultimately up to the private sector to use those tax breaks, and so there are no guarantees, but then who needs guarantees?
LikeLike
In response to your first reply, D would be interest . A negative intertest rate would not occur in the absence of inflation, which is the point of your scheme.
In repsonse to you second reply, if you look at the simpler formula I derived, gives the same r = g result without any additional assumptions. But any market-set value for r will always be greater than g, the difference is the risk premium.
You have the same situation when valuing equity with a discounted earnings growth model. You have to use a discount rate greater than the assumed growth rate in earnings, otherwise infinite valuations are obtained, which sort of defeats the purpose of a valuation analysis–unless you buy into the idea that trees grow to the sky :).
LikeLike
Mike,
“In response to your first reply, D would be interest . A negative intertest rate would not occur in the absence of inflation, which is the point of your scheme.”
No, I was using D to represent the growth of the federal debt level resulting from current deficits. The debt grows from two variables – the rate of interest r, and the current deficit (revenue – expenditures). The point I was trying to make in the first post is that the interest rate can be higher than the rate of growth if the total outstanding debt is falling via federal surpluses. That is why I calculated D to be -6.7%. For a 10% interest rate, and a 3% rate of growth, the federal government would need to run a 6.7% rate of surplus to reach an infinite valuation on all future tax revenues.
Also, the whole point of my scheme was to distinguish between the cost of debt in the private sector and the cost of debt in the public sector (federal government). The private sector realizes its cost of debt service on an after tax basis in many cases (mortgage, student loan, etc.) because the interest payments are tax deductible. I was demonstrating a method in which the after tax cost of debt service in the private sector could be negative in the total absence of inflation (inflation being defined as either a rise in the price level or a rise in the total amount of money / debt).
“In response to you second reply, if you look at the simpler formula I derived, gives the same r = g result without any additional assumptions. But any market-set value for r will always be greater than g, the difference is the risk premium.”
Incorrect. The markets may not set r greater than g for several reasons – risk aversion, lack of access to other markets, legally required purchases, indifference to return on investment, etc. However, the federal government could override market pricing of its own liabilities if it chose to do so.
“You have the same situation when valuing equity with a discounted earnings growth model. You have to use a discount rate greater than the assumed growth rate in earnings, otherwise infinite valuations are obtained, which sort of defeats the purpose of a valuation analysis–unless you buy into the idea that trees grow to the sky.”
Again, incorrect. You have to use a discount rate greater than the risk free rate of return offered by the federal government. That discount rate is often a reflection of credit conditions in the private sector. And so, if credit spreads widen between the federal government and the cost of debt for the company in question, its equity valuation will take a hit.
Infinite valuations are a definite possibility on paper. But there is more to the value of a company than just its capital cost. The company still has to produce something that consumers will buy. It has nothing to do with trees growing to the sky, and a lot to do with market saturation, market competition or lack there of, etc.
LikeLike
Mike,
“In response to you second reply, if you look at the simpler formula I derived, gives the same r = g result without any additional assumptions. But any market-set value for r will always be greater than g, the difference is the risk premium.”
From your original equation:
“For example you suggested 10% for r. GDP growth over the long run is about 3% so we use this for g. Plugging these values in equation 2 gives:
PV =P (1+0.03)/(0.10-0.03) = 14.7 P”
Now if we use the modified equation to include a non-guaranteed rate of return of 10% for r, a GDP growth rate of 3% (actual long term U. S. nominal GDP growth rate is closer to 5.5% over the last 80 years), and a success rate = g – r = -7% we get:
PV = P x (1 + .03) / (.003 – .0001) = -147 P
Notice this is the same as if we use a debt growth variable D = -7% (fiscal surplus is 7%), and the same variables (3% growth and 10% guaranteed rate of interest) we get:
PV = P x (1 + .03) / (0.10 – .07 – .007 – 0.03) = -147P
And so the government’s net fiscal position is improved by the same amount either by running a 7% annual fiscal surplus OR by selling tax breaks that appreciate at 10% annually.
That is honestly something I did not expect. I was kind of hoping you would have put 2 and 2 together to reach the same conclusion after reading both messages.
LikeLike
Mike,
Spotted a mistake in the second equation:
PV = P x (1 + .03) / (.003 – .01) = -147 P
LikeLike
You used D in a discounting formula as one would use an interest rate. If D the deficit, I don’t see the reason for the way you used it.
To make my point clear, let’s look at a simple case as an example. We will use exclusively one year durations and sell reductions of X dollars from next year’s taxes
The price paid would be P= X (1+g)/(1+r)
Lets us further assume that r = g, which gives the infinite case you described above. In this case we get P = X, that is the investors pay a sum equal to their current tax liability for this security.
In the first year of this scheme, the government will receive revenue equal to 2X, this years taxes and forgiveness for next year’s taxes. The government uses some of the extra X to pay down debt, and save some for a rainy day. Next year the government will not collect the (1+r )X worth of taxes it would otherwise get. It collects no taxes at all, but does sell (1+r)X worth of securities like it did last year, so it raises just enough revenue to meet predicted expenses (assume to grow at the same rate g as revenue)
If the budget started out balanced the securities would raise enough money each year to just meet expenses and the government would reap a one-time bonus of X.
But suppose the budget is not balanced. Suppose revenue is less tha expenses by a deficit D that ios structurally built into the budget by tax cuts enacted in the year before this securitization occurred. That the current year and D(1+r) for the next year, and so on. With the deficit, the bonus X will be consumed after less than X/D years, after which either spending must be cut to eliminate the deficit, taxes must be increased, or the government must borrow.
These are the same choices faced in year 0, all the securitization does, even under the most favorable (and I think unrealistic) assumption that r =g, is delay when this choice has to be made. It accomplishes nothing, which is why I still don’t see why it makes sense for the government would do something like this, especially when they can borrow money for less than g (like right now).
LikeLike
Mike,
“You used D in a discounting formula as one would use an interest rate. If D the deficit, I don’t see the reason for the way you used it.”
No, what I did was recognize that the total interest expense paid by the federal government is a function of two variables, the interest rate “r” and the federal surplus / deficit “D”. In the extreme if the federal government is reducing debt at a higher rate than the interest rate (for instance surplus = 15% while interest rate = 10%), then the present value of all taxes goes to infinity even with zero growth. And so the discounting term (1 + r) x (1 + D) is a truer representation of how quickly the debt grows.
From your original equation, if the rate of interest is greater than the growth rate over a significant enough time, then the federal government’s finances become “Ponzi” like. The federal government’s finances become “Ponzi” like when its interest payments consume 100% or more of its revenue. After that point, it is selling new bonds to make the interest payments on existing bonds. But to determine that point you must also include a factor that describes whether the total debt is rising or falling.
Here are three present values of tax revenue (1 for 10% deficit, 1 for balanced budget, 1 for 10% surplus):
10% Deficit – PV = P x (1 + .03) / (.1+ .1 + .01 – .03) = 5.72P
Balanced Budget (Your original equation) – PV = P x (1 + .03) / (.1 + 0 + 0 – .03) = 14.7P
10% Surplus – PV = P x (1 + .03) / (.1 -.1 – .01 – .03) = -25.75P
The present value of all future tax receipts increases as the federal government moves from deficit to surplus – as demonstrated by the above equations.
I then demonstrated that selling non-guaranteed claims on future tax revenue that have a success rate “s” equal to “g – r” is the equivalent of running a surplus.
“With the deficit, the bonus X will be consumed after less than X/D years, after which either spending must be cut to eliminate the deficit, taxes must be increased, or the government must borrow.”
OR, the federal government replaces bonds (guaranteed claims on its revenue) with non-guaranteed claims on its revenue stream. In essence the federal government does a debt to equity conversion.
“It accomplishes nothing, which is why I still don’t see why it makes sense for the government would do something like this, especially when they can borrow money for less than g (like right now).”
It accomplishes a lot. When the federal government sells tax breaks (non-guaranteed claims on future tax revenue) to the private sector, it can lower the after tax cost of debt service in the private sector below zero.
LikeLike