Tax Investment

Michael Blume recently reminded me of a point I made eleven years ago.  At the time it seemed too obvious to be original, but now I’m not so sure.  So let me point out an important negative investment externality:

Raising the long term real interest rate results not only in dropping poor investment projects, whose rates of return are lower than the new rate, it also causally increases the long term rate of return of all better investments, those that would have been undertaken anyway.

This is a huge market failure, due to a lack of property rights in long term investment prospects, as I explain below.  If it were the only investment externality, then free market interest rates would be too low.  This externality is countered, [edit: complicated] however, by a legal failure due to our choosing not to enforce all feasible long term contracts, especially with the dead:

Large legal barriers now hinder us from making deals with the far future, and from saving today to benefit them.  We do not enforce many kinds of terms in wills, and charities are required to spend a certain fraction of their capital each year, to prevent their endowments from growing large.

Given this overall situation, the obvious policy prescription is:

  1. Enforce all feasible long term contracts, and
  2. Artificially raise global interest rates, e.g., tax investment.

The first policy can be implemented locally, but second seems to require global coordination – an extra local tax may on net hurt that locality.   So without a world government we may never implement this second policy.

Regarding the source of the problem, here is my old argument:

A lack of long-term property rights in most investment projects means that returns above the market rate are burned up an a race to be first.

There are, in fact, very few long-term property rights regarding the right to undertake investment projects. Think of developing a new kind of car, colonizing the moon, developing specialized CAD software, or making a movie with a certain kind of gimmick. Each such project requires various forms of capital, such as machines or skilled labor. While in the short term only one investor may have the rights to tackle a given project, in the long term many competing investors could have positioned themselves to have this short-term opportunity.

For example, Microsoft’s dominant position in PC operating systems now gives it the right to many very attractive investments. But there was once an open race to become the dominant operating system, and competitors then tried harder because of the prospect of later high returns. And when deciding whether to enter this earlier race and how hard to try, investors mainly wondered if they could get a competitive rate of return. Similarly, while one group now has the right to make the next Batman movie sequel, there have long been open contests to create popular movie series, and popular comic strips.

Consider a typical as-yet-untried investment project, becoming more and more attractive with time as technology improves and the world market grows larger. If there wasn’t much point in attempting such a project very long after other teams tried, then a race to be first should make sure the project is attempted near when investors first expect such attempts to produce a competitive rate of return. This should happen even if the project returns would be much greater if everyone waited longer. The extra value the project would have if everyone waited is burned up in the race to do it first.

Thus most of the return above the market return … should be burned up, leaving the average return at about the market return.

Of course some of the wasted effort to invest too early will produce positive innovation externalities.  But it seems wishful thinking to expect this to completely negate the earliness waste.

GD Star Rating
Tagged as: , ,
Trackback URL:
  • Mike

    I’m reading this without an economics background.

    But, it seems to me, if everyone adopted the strategy of waiting for every investment opportunity to produce a higher rate of return, none of them ever would because this progression relies on exploiting external innovations that wouldn’t have occured.

    In other words, it seems to me the “positive innovation externalities” that you refer to as “waste” in the second last sentence, are what drive the tendency for investments to have a higher rate of return if only they were delayed.

    I guess the more accurate statement would be this: if you consider a single investment in isolation, ideally there is some curve that gives the rate of return as a function of the time of initial investment. As you say, people invest when the rate of return in competitive, but maybe they should cooperate in some way and wait for it to get higher. But, if we now allow everyone to act this way, on all investment projects, the slope of this rate-of-return vs time curve decreases.

    Perhaps your last sentence is to say there is no reason to believe it decreases to zero, meaning there is still some advantage to investors to wait.

  • Robin Hanson

    Mike, for each investment, the rate of return as a function of time usually has a peak. Waiting for this peak doesn’t mean you’d wait forever, so investments would still happen and still provide whatever positive externalities they can.

  • Michael Sullivan

    Some more concrete examples of what you are talking about might help. Right now, what you are suggesting seems obviously wrong.

    I don’t see how raising the interest rate positively affects the market return of all better investments. It does positively affect the rate at which you can earn rent on monetizable capital, which is really good for the haves but not so good for the have nots, and should be a wash for the economy as a whole in money terms.

    I don’t see how the race to be first represents a general externality either. It is certainly a negative for companies who don’t end up being first, but getting access sooner is a benefit to consumers. And getting a greater return from waiting, seems also to be a wash economically. It sounds like you are talking about more first mover rent.

    Similarly, what kind of rights are you talking about with long term property rights? I’m envisioning some kind of patent in perpetuity, like the eternal copyright we seem to be headed toward, and it’s quite hard to imagine how that could possibly create more wealth for the economy as a whole. It’s easy to see how it would create more wealth for whoever is savvy enough to acquire the appropriate rights when they are cheap, though. But it would come at the expense of other potential innovators and consumers, as far as I can tell, so it’s not at all clear how this represents a market failure.

    • Patri Friedman

      Yeah, it is hard to understand without examples, I wouldn’t have understood the post if I hadn’t already worked through it on my own.

      The negative externality is on who would have discovered it second. Consider a drug that can be discovered in 9 years of R&D for $110M, or in 10 years of R&D for $10M. Whoever discovers it first will patent it and get $120M in revenue out of $200M in total social benefit. Obviously it is worth spending $110M (an extra $100M) to be first, to get the drug a year earlier.

      There is no reason to think that society values that extra year at $100M. It is just a result of the “first one gets the value” structure. 1/2 of the $200M in social profit is being burned by this structure. Rather than a societal surplus of $200M-$110M = $90M, we could have had a surplus of $200M – $10M = $190M, for only a 1 year delay in getting the drug.

      • Michael Sullivan

        So, it makes more sense if I think of the long-term rights that solve the problem not as extended patents — because the problem is the race to be patentable. What you want is a lower barrier to lock up research toward a particular patentable item early on in the dev process. So you could apply for a pre-patent right to deliverable a patentable item in N years. This would then be made public and bid on. Like other patent rights it should be tradeable. What the winner (holder) gets is basically the sole right to patent this item if they produce it within the term. If anyone else wants to beat them out, they have to buy the pre-patent or hope the holder can’t produce a patentable product within the term so that it expires.

        If this is the kind of right Robin imagines, that would make sense (at least in the patent space). There is still the social cost of the year wait, and while there’s no reason to think it’s as much as 100 million, it’s still significant (probably on the order of 20M if the net benefit over the patent term is 200M).

        Plus we haven’t considered how much of the extra 100M in spending is dead-weight loss versus mere utility transfers. In your contrived hypothetical, it’s easy to imagine dead-weight loss being more than 20%, so that probably represents a net social cost, that some kind of pre-patent right would eliminate or at least drastically reduce.

        I’d love to have a real world example to chew on, though. I’m not aware of any races to be first where the cost ratio was that dramatic (10 x cost for 10% time savings). With less extreme numbers and a small percentage of dead-weight loss on the extra spending, a race to be first in innovations could end up being a net social positive, even if it is negative for all the entrants in the race.

        The social benefit is less obvious for something like homesteading.

        I’ve thought some more on the interest rate issue, and if there is no economic profit (due to efficient competition), then it makes sense that all investments would earn the market rate plus something the investor brings to the table. But the difference is still just transfers except at the investment margin which, as Jordan Amdahl notes, is an unclear (to me anyway) trade off between delayed now profitable investments (social good) and scrapped no-longer profitable investments (social bad).

  • Douglas Knight

    I found some of the phrasing confusing. Using “long term rights” to mean monopolies is accurate, but confusing because of the connotations. In particular, it primes us to think you want to create the rights. (Maybe you do, but it’s distracting from the main point.) That is a mistake both Michael Sullivan and I made. Also, the blog post combines that with mention of long term contracts. That made me jump to the conclusion that you thought long term contracts could enable the long term rights.

  • Jess Riedel

    I was also confused with the phrasing. Would it be accurate to summarize the discussion of the long-term market failure as “Patents have a finite lifetime of 20 years, so investments in technology with a time horizon longer than 20 years will be underfunded” ?

  • Robin Hanson

    Apparently I should only post on economics topics to a special blog that only professional economists can read. The rest of you should not expect that random amateur exposure prepares you to grasp all possible econ insights.

    Jess, this applies equally to the patentable and the unpatentable, to short and to long term investments. Douglas, I’m sorry you have misleading connotations. Michael I gave several concrete examples; I can see that you can’t see; you’ll just have to learn more econ.

    • Jess Riedel

      Robin, I don’t think your condescension is justified. No one suggested that “random amateur exposure prepares you to grasp all possible econ insights”. Us amateurs were genuinely confused and we asked for help. No, you don’t have to give a mini-lecture in response to every elementary econ question. It’s a waste of your training. But that doesn’t mean you should interpret honest questions as insults. Why not just let someone else answer them?

    • TGGP

      You’re right that non-economists aren’t going to understand much of what you have to say on economics, but a lot of the more accessible stuff also draws on economics. The title of this post reminded me of “tax the tall”, which was widely discussed in blogs by non-economists (although Greg Mankiw was more the instigator of that).

    • Mike

      Even though I’m one such “amateur,” I think I learn a lot from reading this blog, and I’ve learned much from the occasional replies to my comments from you and others. (Thanks!)

      I realize these replies are in essence an act of charity, so in the future (as I did above) I will briefly advertise my “amateur” status at the very beginning, so you and others can choose whether to spend time on my comments or not (no hard feelings if not — we’re all busy people). Maybe we should all do that, which has the side benefit of allowing us amateurs to discern the validity of those comments that receive no reply.

    • Patri Friedman

      I think the problem is that you didn’t use any concrete examples. I have been able to explain a simpler version of the same argument (which does not use interest rates / rates of return) to many people.

  • gwern

    In the linked piece, the Dyson Wired link is dead; the new URL is

  • Agent00yak

    A few points: (Is the joke that if you only want responses from non-economists then you would publish your ideas in a gated journal?)

    A. It looks like you are largely discounting the investment vs consumption trade off by having time preference being dependent mainly on genes. This doesn’t seem right. You might believe that the race to be first has negative externalities that are larger than the foregone investment at all in most cases, but that seems unlikely.

    B. How much of the benefit comes from proposal #1 and how much from proposal #2? Proposal #1 alone seems beneficial, proposal #2 alone seems harmful as it will drive people towards even shorter term behavior. #1 and #2 might be better than the status quo, but how much of that is from the #1 and how much from #2?

    C. The economy reacts to the price signals of bottlenecks, and perhaps this reaction is slow but raising the rate of return to get around the bottlenecks gives far too much credit to the foresight of actual economic actors, who are more responsive to current price signals than possible future price signals.

    D. Short term investments in the stock market are already taxed at a higher rate than medium term investments. Perhaps the term structure of taxes could be a better mechanism to theoretically modify than than the total level of taxes on investment.

  • Jordan Amdahl

    I’m still digesting, but a couple comments:

    1. “Poor” investment projects discouraged by artificially higher interest rates is not a good outcome. Investments that are “poor” at the margin still are a net benefit to society. The increased profitability of investments that are delayed is likely to be offset by investments that never happen. While it might be true that these investments are only marginally profitable, a marginal increase in the interest rate will only make the delayed investments marginally more profitable. It is not at all clear to me that there is a net gain to be had.

    2. The use of tax policy is problematic because it will lead to many unintended consequences. For example, investments tied to more secure property rights will only be adversely affected. For example, suppose I own a natural resource. I can rationally choose to invest in extraction when the profitability of doing so it at its highest. Even if the maximum profitability occurs after my death, I can wait for the resource to appreciate as technological advances make it more profitable and sell it to someone who will live long enough to invest in it (or, more likely, to a corporation). In such cases, there is no tragedy of the commons where investment is made too early and taxes on investment will only lead to deadweight loss.

    • Robin Hanson

      On your point 1, it is exactly like why a monopolist wants to raise prices above the competitive level; he loses marginal customers and gains on all infra-marginal customers.

  • Constant

    This reminds me of:

    The homesteading problem–inefficiently early settlement in order to establish a claim.

    • Robin Hanson

      Yes, that is a more specific example.

  • Gary

    If there wasn’t much point in attempting such a project very long after other teams tried…

    Firms can’t compete in the same technology space if the first-mover secures those rights, but they can compete above that technology space as the tech becomes available. No problem here.

    This should happen even if the project returns would be much greater if everyone waited longer.

    The returns to using the first tech that satisfied “market return + epsilon” couldn’t possibly be greater by waiting and then still using that same tech, so what do you mean here? If one firm lunges after the “market return + epsilon” opportunity, other firms will go after the new tech opportunities as they become available. All opportunities are captured.

    Do you mean that firms should collectively ignore the current tech that satisfies “market return + epsilon”? Why? It seems to me that firms that bring a recently profitable opportunity to market satisfy a market demand from consumers who are saying, “I will buy your product even though other firms are bringing new and better technology soon”? Why second-guess their time-preferences?

  • Tim Tyler

    I think the problem with this post is that it apparently claims that raising interest rates is good without reference to the current rate (which everyone who has heard of Zimbabwe knows is wrong) and then defends the thesis using inpenetrable economist tech-talk. That sort of thing is bound to make many readers squirm.

  • Robert Bloomfield

    I am with the folks who are confused by this post, and probably don’t qualify as an amateur. I would suggest breaking this post up into a few separate, more specific arguments.

    If we move away from macroeconomic policy, does the argument have anything to say about the following? Private firms often demand new projects to generate returns that far exceed their cost of capital. They tout this under the banner of ‘capital rationing’ and traditional economists have struggled to understand the policy. Is there a relationship?

  • fburnaby

    I’m not convinced that Robin’s choosing to neglect us amateurs is necessarily in his interest (though of course, he’s free to). When I program, I sometimes explain how my code works to this or that inanimate object, which sits on my desk. I have to say things very carefully and clearly in order for the “stupid” thing to understand me. I tend to find most of my bugs this way.

    Maybe some time, when you get the chance, you could try to help make your discussion on this topic more “accessible”? I know I’d enjoy reading it in words I can understand more readily, and if your theory is truly as novel as you claimed it might be in the intro to your article, then you’ll probably also benefit from an increased reader-base (assuming that’s something that appeals to you). In any case, may help Robin extend his own understanding of the implications by having such a discussion.

    This is probably the hippiest free love post I’ve ever made on a blog. Peace and love all around! ~

  • Patri Friedman

    I’ve been making the same general point – that people burn resources in an attempt to be first, and that this may be a very common situation where individual rationality leads to bad group outcomes – for years, ever since I first encountered it. I hadn’t thought of long-term property rights in opportunities as an answer.

    Also, I had focused on it in narrower contexts where the problem is more obvious, like patenting drug discoveries, or discovering stock market mispricings. I find that using such narrow contexts helps people understand it. I think of it as a form of rent-seeking, burning most of the value of a windfall or other fixed resource by competing to obtain it. What I am not sure of is how widespread it is in practice.

  • deo

    Robin, if I understand you correctly, you suggest, that by delaying a project until technology improves, so that project’s rate of return increases, you can increase average rate of return, and thus, growth rate.

    But in your model you take that demand schedule moving up is completely exogenous process, having nothing to do with actual projects being run through trial-and-error process. This doesn’t seem right. Of course, you mention positive innovation externalities and assert they won’t outweight losses due to first-past-the-post race. But it would be nice to provide some evidence for this assertion. Can you point some market where judicious delaying of well-thought-of projects actually provide superior returns and doesn’t, at the same time, slow down rate with which demand schedule goes up?

  • Constant

    This entry appears to presuppose the “first mover” hypothesis made infamous by the dot-com crash.

    Of course, another reason people attempt to be first (the attempt to be first being the culprit here) is that the first person with an innovation gets to patent it. But the solution may simply be to abolish patents.