Economics and similar, for the sleep-deprived
A subtle change has been made to the comments links, so they no longer pop up. Does this in any way help with the problem about comments not appearing on permalinked posts, readers?
Update: seemingly not
Update: Oh yeah!
Tuesday, January 07, 2003
This is dividend, my friend, my beautiful friend
I like to argue things both ways. Here's an example:
1. It is obvious that the effect of the Bush proposals on dividends will be to increase investment. Cutting taxes on dividends makes stocks more attractive to own, which encourages investors to put their money back into the market.
2. It is obvious that the Bush tax proposals will have a negative effect on investment. When you reduce the tax on something, you increase the incentive to do that thing. If you decrease the tax on dividends, you encourage companies to pay out more of their earnings in dividends, leaving less retained earnings for them to invest.
Which is right? Well, ask an economist, get a rambling answer which leaves you just as confused as you were at the beginning but somehow manages to work its way round to a tirade about personal hobby-horses. Suffice it to say that anyone who was hoping that Post-Keynesian economics weren't going to be at the root of this one is going to be disappointed.
First up, a public service announcement. In an entirely understandable rush to bash Bush, a lot of my coreligionists appear to be saying some pretty silly things about dividend taxation, including a few, like Paul Krugman, who really ought to know better (MaxSpeak does it about the best). In ascending order of objectionableness to me:
- It's amazingly inegalitarian. Actually a fairly good objection; obviously a tax cut on dividend income is targeted, as Krugman says "like a laser" on the very rich. But this is America we're talking about, not Sweden. If you're going to complain about inegalitarianism in public policy, you might not want to start here. And in any case, to concentrate on single proposals is bad analysis and bad policy; it's the progressivity or regressivity of the total stimulus package that matters.
- It won't stimulate the economy. There could be a sensible version of this critique, but it would have to be hidden in so many thickets of caveats it wouldn't be worth making, and I don't think that Max's argument (that dividend cuts will be offset by the sales of bonds to the same investors) is convincing. Some of the money from the dividend cut will go straight into the government bond market, sure, but some of it will be spent, and some of the deficit will be financed by foreigners. The plain facts of the matter are that a reduction in tax on individuals, gives money to individuals and that's stimulative. It might not be the most efficient way in which to use a deficit of given size to stimulate the economy (in all honesty, the current plan looks like a load of economists got together and said "Hey! Let's smoke a really big rock of crack and design a stimulus package"), but I don't think that one can deny that it is a measure which, in principle, could have a stimulative effect on the economy.
- The argument that changing dividend taxes removes double taxation is in some way mendacious or misleading. It isn't. There's a plain fact of the matter here. You pay tax on dividend income, and corporations pay dividends out of post-tax profits. Hence, two levels of tax. One might argue that double taxation is the price one pays to enjoy the benefits of limited liability, and that would be a fairly decent argument. What isn't a fairly decent argument is to start talking about sales tax or what have you. That's a clearly misleading analogy; if you were to count the tax you pay when you spend the money, then dividends are taxed three times, not two.
So anyway. What's to recommend altering the taxation of dividends? Well, quite a bit, on general taxation policy grounds. Obviously, the Bush team are, to put it as charitably as possible, overly keen on structural solutions to cyclical problems (to put it more bluntly, a la Krugman, they're keen on structural non-solutions that make cyclical problems worse but benefit some very narrow class interests). But altering the system of dividend taxation in this way is not a bad thing to do.
Why? Well, the point is that debt interest is a tax deductible expense for corporations, but dividends aren't. So (and here I oversimplify mightily), someone planning the financing of his company is comparing the pre tax cost of debt with the post tax cost of equity. It makes a big difference in dealing with the investors and capitalists who finance your plant and equipment, if you give them something that you can pay out of pre-tax income; effectively, debt financing is subsidised to the tune of the rate of corporation tax. This is a subsidy to gearing, and as any ex-employees of Enron or Global Crossing will know, debt gearing is not necessarily something which the government ought to be in the business of subsidising. As well as the general principle of tax policy that a tax ought to cause as few distortions in behaviour as possible, subsidising gearing is particularly pernicious because it fuels booms and busts; it's a lot easier to get a dodgy takeover, leveraged buyout or startup proposition off the ground if you're always able to make the post-tax cash flow numbers look a bit better by adding more debt to the balance sheet. So a subsidy to debt is a subsidy to the creation of WorldCom-like companies, with balance sheets that are incredibly vulnerable to any change in sentiment or business conditions.
But under the proposed Bush plan, the idea is that every time a company pays a dividend, it passes out along with it a tax credit, allowing the investor to deduct the corporation tax already paid on the profits from which the dividend derived from his income tax bill. It's a lot more complicated than that, obviously. But that's the gist of the thing. In principle, and if it works, there will no longer be an incentive to gear up on debt, because investors will be prepared to provide equity capital on cheaper terms in order to get those lovely tax credits.
Which is where we came in. There's clearly two factors at work here; on the one hand, dividend tax "reform" encourages people to provide equity finance to corporations on cheaper terms. But on the other hand, having provided that finance, it encourages them to demand that the profits of those firms are paid out as dividends rather than being reinvested. What's the net effect?
I'm going to ignore a whole strand of the financial economics literature here which suggests that the demand for larger dividends is irrelevant. Basically this theory, which quite deservedly won Franco Modigliani a Nobel Prize, points out that a company which pays out its entire profits as dividends can always then ask its shareholders to subscribe to a new equity issue to fund any profitable investment projects it has lying around. This theory actually goes back even further to Irving Fisher (who proved that investment decisions in capital are logically separable from decisions about how to finance that capital), and goes forward in time to Michael Jensen, who argued that it's actually better for firms to pay more of their cash flow out, because the discipline of the marketplace is better at making investment decisions than incumbent company management. I'm going to ignore this strand of the literature because a) empirically, firms didn't behave in this manner in the UK during the period in which we abolished double taxation of dividends, and Stuart Myers gives decent theoretical reasons why (asymmetric information between lenders and borrowers means that internally generated cash is always the cheapest source of income), and b) Michael Jensen was the academic guru of leveraged buyouts and junk bonds, so I regard it as safe to laugh at him rather than arguing against him these days. Terribly unfair I know.
But the Post-Keynesian point I want to make is not that far removed from Fisher's theorem proving the separability of investment and financing decisions. The point is, as Paul Davidson spends the first five chapters of his excellent book "Financial Markets, Money and the Real World" (no amazon link provided as the bloody thing costs eighty quid!), the word "investment" in the modern world has two meanings:
a) "Investment" refers to the purchase of capital goods by entrepreneurial firms
b) "Investment" refers to the purchase of financial securities by individuals.
What do these two definitions have to do with each other? According to the Post-Keynesians, not enough.
There is no very strong force at all which equilibrates Investmenta, the amount of money that firms want to spend on capital goods with Investment b, the amount of money that you and I want to chuck into the financial markets. In national income accounting, they have to add up, but that's a bit of a cheat; national income accounting counts stocks of unwanted goods piling up in a warehouse as "investment", and in this sense, ex post, investment has to be equal to ex post spending. But it's clear that this sort of "investment" isn't going to drag us out of a recession.
Then there is a weak force trying to bring the two into equilibrium, which is the same sort of ambiguous, difficult-to-understand kind of supply/demand dynamics which apparently tells that that government deficits are (on the whole and in the long run) associated with higher interest rates (on the whole and in the long run). If there's a lot of spare investmentb hanging around, then it will tend to depress the return on investment, which lowers the opportunity cost of investmenta, because you might as well buy a new lathe if all you can get in the stock market is 3% a year.
But this equilibrating force is very weak. The discount rate (the opportunity cost of capital tied up in a project) is important in capital budgeting. But it's not the most important thing. If you're planning on a bit of capital investment, the most important thing you want to know is whether there's going to be a demand for the goods you want to produce at a price which will keep you in business. How do you work out the answer to that question? Anyone? Bueller?
"Sir, you can't work out the answer to that question in any statistically reliable manner, sir, because economic processes are nonergodic, sir, because the economy is subject to positive destabilising feedback sir!"
Very well done that pupil. The rest of you, go back to the books and revise.
And now we have most of the theoretical framework to understand Keynes' great insight, Post-Keynesian interpretation thereof. The great problem in the economy is in managing the fluctuations in investmenta, a quantity which is not directly related to any economic variable today in any straightforward way, with investmentb, the planned saving of the economy, because this is the level of saving at which the economy can be in full employment equilibrium. An excess of investmentb over investmenta is what is meant by "excess saving" or "deficient aggregate demand".
To take this further, we need to think exactly what we mean by "saving" in this context. The reason that investmentb is called "investment" in the first place is that a lot of what people do when they save involves buying claims on capital assets; long term debt or equities. Although there can certainly be an excess of this kind of saving relative to capital investment plans, there is another important distinction to make which gives the model its specifically Keynesian flavour. This is the distinction between the personal sector's decision to hold claims on capital assets, and its desire to hold money balances.
Money is at the centre of Keynesian theory, because it is quite unlike any other savings instrument. Specifically, it's fully liquid, and it isn't invested in anything. You own money balances because you want something that can be converted into consumption goods instantaneously and with certain value. Some readers may remember this discussion from my now-restored archives on the subject of risk-free assets; the point I was rather obliquely trying to make was this one; for some purposes, particularly the postponement of a fixed amount of consumption in the abstract from today into an unspecified point in the future, nothing will do except a liquid asset. And the point about a liquid asset is that it's not invested in anything; if it was, it would be illiquid.
Which means that the composition of savings matters, because that's what determines whether the weak equilibrating force can work at all. Money is not a substitute for securities for people who care about liquidity, so it is possible for world to get into a situation in which the saving population wants to hold money and doesn't want illiquid assets (because asset returns are also assumed to be nonergodic, and thus there is no scientific way of answering the question "at what level of the Dow am I fairly compensated for the risk?"). In this situation, because money is not a substitute for illiquid assets, there is no change in the price of illiquid assets which will convince the population to buy illiquid assets (and thus finance investmenta) rather than holding money, and whatever money is poured into the system will be held rather than spent on illiquid assets. This is the "liquidity trap".
So to recap, there are two distinct Keynesian concepts of a recession. You can have a situation in which the companies who carry out investmenta don't want to invest because they don't expect the demand to be there for their products, whatever the supply of capital. And you can have a situation in which savers don't want to carry out investmentb because they don't expect any investmenta plans to be successful, whatever the demand for capital. Note that in both cases, the imbalance turns into a problem because a proportion of the population is employed in producing goods the consumption of which is unrelated to current income, because they are capital goods whose utility depends on perceptions of future income and consumption. If those perceptions get out of line in the economy, then the market for capital goods has some buyers who either can't (for lack of finance) or don't want to buy them, and there will be a negative income shock to the producers of those capital goods which is not offset by a positive income shock elsewhere.
So, is the Bush plan stimulative? Well, it frankly depends on what sort of a recession we are looking at. The plan under discussion would subsidise the supply of capital, by reducing a tax on the activity of investmentb. That might conceivably help matters if we were in a situation in which the problem was a liquidity trap; it's effectively a subsidy from the government which you only get if you invest in illiquid assets. But I personally don't believe that this is a reasonable characterisation of the US or world economy at the moment; I don't think that what is holding the economy back is a shortage of financing. If the problem in the economy is that the corporate sector is unable to digest the last big investment it made and can't see any profitable new investments at all, then reducing the cost of capital isn't necessarily going to help matters. The prescription for that kind of situation would be to do something to stimulate demand. Which we've already established that a dividend cut will do, but probably not enough ....
Endnote: The probability that I've got those superscript a's and b's right is pretty small. If anyone notices a wrong 'un, give us a shout in the comments section?
this item posted by the management 1/07/2003 09:12:00 AM