Could someone explain Bob Murphy’s ‘monetary’ theory of interest? I have a decent familiarity with PTP, and I’m listening to his lecture on this, but I’m not quite clear of either his critique or his alternative. Also any articles other authors have written on Murphy’s theory in this regard.
I have had some thoughts about a medium of exchange being perhaps necessary in order to establish a regular intertemporal rate of interest, but I haven’t really made any effort to work it out.
I found that lecture somewhat perplexing, its sort of rambling and aimless. And from someone that’s written study guide to Human Action, I remember finding some of his statements rather odd. Interest is a universal phenomenon … so of course you can theorise about money and interest but you can theorise about interest without considering money (even if money is present in advanced societies featuring economic calculation)
Well, what I was sort of thinking is that the heterogenaity of intertemporal goods might make it problematic to arbitrage into a uniform rate of interest unless you have a money medium of exchange; because it’s not clear how you would establish the ratio of scarcity otherwise to set or even recognize the opportunity.
I have no idea if that’s what Murphy is talking about, though.
Yes, there can only be a uniform rate of interest in the impossible evenly rotating economy, not in the real world. This is laid out plainly in Human Action.
I’m taking his class, and a couple times he’s come off like this. He does eventually get around to explaining things in a very sensical way, in the end.
What I’m saying is, I’d bet he has a good idea in his mind what he wants to present and talk about, he’s just not presenting it in the best manner. I’ll take a look at the lecture and try to make sense of it…
And no, his instruction isn’t terrible. At its worst, it’s still good. 
I’m taking his class, and a couple times he’s come off like this. He does eventually get around to explaining things in a very sensical way, in the end.
What I’m saying is, I’d bet he has a good idea in his mind what he wants to present and talk about, he’s just not presenting it in the best manner. I’ll take a look at the lecture and try to make sense of it…
And no, his instruction isn’t terrible. At its worst, it’s still good.
I can understand having difficulty explaining to other people just what you mean and going on extended tangents, I am exactly the same way. I enjoy reading his blog, I just never got the point of his lecture.
2 interesting points I think he made in that lecture were a statement about uncertainty he quoted from Karl Popper and the fact that PTPT made use of interpersonal utillity comparisons, between future and present individuals to explain interest in “exceptional cases”(e.g. a kid preferring to leave a beer to drink when he is older, than now).
I sympathise with his former statement, I feel David Gordon made a weak counter argument against it during the questions following the lecture.
On the 2nd point; after thinking about it, I figured that PTPT does not commit this kind of mistake, since it ascribes the comparison as made by the acting individual at one time before his action. The “interpersonal” utillity comparison made by the person upon acting, in “estimating” the future utility he will likely derive from a certain good may well be unscientific, though this does not affect the evaluation of his actions based on such an appraisement and subsequent valuation by us as praxeologists.
Though perhaps I’m talking nonsense…
This is essentially Sraffa’s critique but it ignores Bohm-Bawerk’s major theoretical contribution in TPTC, namely that the rate of interest (natural interest) is an entirely independent economic phenomenon, which exists whether it is correctly expressed in capital markets or not.
I think Murphy is essentially saying that we should ignore “natural interest” because it’s intangible, and that we should solely focus on the monetary element. But this would throw out the ABCT and Austrian capital theory entirely, reducing it to the Keynesian framework. I think Murphy’s confusion stems from the fact that he doesn’t fully appreciate the implications of Mises’ (Knies’) three-fold distinction of goods into producer, consumer, and media of exchange. There is a demand for money as money (for transactions and liquidity) and it is both a class and a good in itself. Thus the monetary variable and the “real” variable, so to speak, both influence the rates of interest which are expressed in the capital and money markets, and we shouldn’t ignore asymmetric information and the element of risk (which stems from differing time horizons on maturaties).
I think he was being vague on purpose. He realizes the implications of his theory.
I understand that with time preference we are going to have interest rates on borrowing, but this is going to be distributed among varying individual time preferences and basically incomennsurable intertemporal goods. I can’t see how you would get a uniform interest rate, or anything approaching a ‘rate’ of interest (rather than a unique price on every intertemporal exchange depending on the time preference of the two interactors) without money prices.
Arbitrage. Each phase of the economy, both vertical and horizontal, will equalize the various interest rates into one uniform rate based on the time preferences of society (expressed by the ratio of demand between consumer and producer goods, aka saving/consumption rates). Of course, this assumes non monetary influences and perfect information. The introduction of money allows this rate to be expressed as a price (can’t have observable prices without money), but then introduces a new set of problems.
Again, the interest rate is the price of time in the general sense. You have to remember that no two goods are truly homogenous; two physically identical apples are heterogeneous, and yet arbitrage will equalize its market price (objective exchange value) when there is perfect information.
But how can you arbitrage without money prices? There is no way to compare all the various exchange rates of present goods, much less future goods, without money.
Of course, this assumes non monetary influences and perfect information.
and because these assumptions are not reasonable but rather patently absurd, the thing that they have been assumed for is exposed as untenable… right? We say that there is a tendency towards uniformity/equilibrium that is never reached etc.
I imagine it would be rather difficult to do since rational calculation is impossible without money prices, but this does not damage the theoretical tenability of “natural interest.” It’s what the interest rate would be when money exerts no active influence and when all information is given. It’s the market clearing price for capital goods. It cannot be expressed as a price when there is no money (like everything else), and we should not expect the market rate and the natural rate to be equal, but that doesn’t mean that it doesn’t exist, or that the concept is entirely useless. The Austrians deduced their business cycle theory from it.
Sraffa attacks it, I think, because he cannot deal with the element of time in his purely static, and quite frankly, ridiculous framework (it also invalidates his trade cycle theory).
The various points of Murphy’s argument(s) seem to be as follows:
Positive time-preference is not is not a praxeological law i.e. not necessarily true a priori.
The Austrian “pure” positive time-preference theory of interest (PTPT) is implicitly based on cardinal utility (as it holds subjective preferences constant and looks at intertemporal value ratios), and is thus inconsistent with Austrian praxeology.
3.1. The PTPT is not reflective of how interest functions in the real world or even in a dynamic economic model, and is neither necessary nor sufficient for an explanation of interest (for example, because negative interest rates exist). (he’s right)
3.2. In order to explain the actual function of interest in the real world or in dynamic models, the theory of interest should incorporate other factors such as the marginal productivity of capital (e.g. classical theory of interest) and the role of money as something that leads to positive interest rates via marginal desire for liquidity in the face of future uncertainty. Money is central to the particulars of intertemporal exchange and therefore a theory of interest that takes the special role of money out of the picture (eg. PTPT) is problematic given that interest is a nominal rate, and money by definition is characterized as interpersonal media of exchange and intertermporal store of value.
I think he’s mostly right, and I agree with esuric that this has some pretty significant implications for Austrian theory. I think it may tend to the conclusion that given a state monopoly on currency, macroeconomic theories that deal with monetary disequilibrium (eg. Keynesian) have some conditional validity. If anyone is more familiar with the various theories of interest, I’d be interested in an analysis of the implications of a more monetary theory of interest for macroeconomics.
I think his stuff about the deductive validity of PTPT vs. the real world or dynamic model application speaks to a more general problem with Austrian method (primary a priori deduction) as it is epistemologically limiting in a way that empirical study is not. I think if Austrians incorporated these implications into their method, they’d just be very consistent neoclassical microeconomists. It may throw off ABCT, but not necessarily its conclusions.
- Positive time-preference is not is not a praxeological law i.e. not necessarily true a priori.
This seems nonsensical to me. Ceteris paribus, an agent perfers present AND future satisfactions, thus time prefernce must be positive.
Ceteris paribus, an agent perfers present AND future satisfactions, thus time prefernce must be positive.
Time preference has to do with the relationship between present and future satisfactions for consumption of thing X, specifically the difference between those satisfactions for X intertemporally. There are good intuitive and inductive arguments as to why time preference trends strongly positive for most things, but it’s not necessarily true a priori.
Hi Cal, what do you make of this ?
The Foundations of Time Preference Time preference is implied in the very nature of valua- tion, and, indeed, of human life itself. All other things being equal, to want something is to want it sooner rather than later. If all other things are equal in two succeeding periods of time and a good exists which could be consumed in either period, then the very fact of the good’s being valued implies that it must be consumed in the first period. If it is not consumed in the first period, then the identity of conditions implies that it also cannot be con- sumed in the second period. Hence, the good simply would not be consumed and, by implication, its con- sumption would be demonstrated not to be valued. If, however, the good is consumed in the first period, its nonconsumption in the second period does not contradict its being wanted just as much in the second period; it is simply unavailable in the second period. The nature of human life implies time preference, because life cannot be interrupted. To be alive two years from now, one must be alive one year from now. To be alive tomorrow, one must be alive today. Whatever value or importance one attaches to being alive in the future, one must attach to being alive in the present, because being alive in the present is the indispensable precondi- tion to being alive in the future. The value of life in the present thus carries with it whatever value one attaches to life in the future, of which it is the precondition, plus whatever value one attaches to life in the present for its own sake. In the nature of being alive, it is thus more important to be alive now than at any other, succeeding time, and more important to be alive in each moment of the nearer future than in each moment of the more remote future. If, for example, a person can project being alive for the next thirty years, say, then the value he attaches to being alive in the coming year carries with it whatever value he attaches to being alive in the following twenty- nine years, plus whatever value he attaches to being alive in the coming year for its own sake. This is necessarily a greater value than he attaches to being alive in the year starting next year. Similarly, the value he attaches to being alive from next year on is greater than the value he attaches to being alive starting two years from now, for it subsumes the latter value and represents that of an additional year besides.The greater importance of life in the nearer future is what underlies the greater importance of goods in the nearer future and the perspective-like diminution in the value we attach to goods available in successively more remote periods of the future.
All other things being equal, to want something is to want it sooner rather than later. If all other things are equal in two succeeding periods of time and a good exists which could be consumed in either period, then the very fact of the good’s being valued implies that it must be consumed in the first period. If it is not consumed in the first period, then the identity of conditions implies that it also cannot be con- sumed in the second period. Hence, the good simply would not be consumed and, by implication, its con- sumption would be demonstrated not to be valued. If, however, the good is consumed in the first period, its nonconsumption in the second period does not contradict its being wanted just as much in the second period; it is simply unavailable in the second period
This appears to me to be nonsense. Setting “all other things equal,” there is no reason why “the very fact of the good’s being valued implies that it must be consumed in the first period.” Remember we are keeping good X constant through time here and comparing the subjective valuation of consumption of that good now vs. some time deferred into the future. That’s the whole point. It’s not that the economic agent may himself at a future “now” value good X more or less or equally. It’s about valuing Xconsumed now vs. Xconsumed in future. It’s “time preference” not “preferences through time” (though the former may be subjectively formed by prediction of the latter).
Murphy correctly points out that holding good X constant means essentially that X is materially interchangeable i.e. the consumer will not notice a difference with X if we swap Xnow with Xfuture. Thus, if this paragraph is arguing that the “identity of conditions” of consumption of X change through time, then it is missing the point of “time preference.” For example, if Ludwig during the winter (now = winter) values consumption of ice (X) less than the consumption of the same ice in the summer (future), then the value of future consumption of Xfuture > consumption of Xnow. And thus positive time preference is not a praxeological law. Again, if the Austrian then argues that it is not the same good because the "conditions of consumption have changed, then he’s missing the whole point of “time preference.” Conditions are always changing. That’s what time does.
The greater importance of life in the nearer future is what underlies the greater importance of goods in the nearer future and the perspective-like diminution in the value we attach to goods available in successively more remote periods of the future.
This is intuitively true, generally speaking, and a reasonable assumption to make probabilistically for most goods. It would also be easily empirically testable. It is not a praxeological law.
Anyway, the thrust of Murphy’s critique doesn’t depend on there being negative time preference, it’s just a sidenote to his central critique.
Ice-in-the-winter versus ice-in-the-snow is not keeping the good constant.
what you can do is analyse ice-in-early-winter versus ice-in-late-winter keeping all things constant, and you can separately analyse ice-in-early-summer with ice-in-late-summer.
you will see that all else being equal (i.e its ice in the winter whether early or late), if the person prefers the ice later rather than sooner, then this will be true at any point in time of the continuum through winter, since that whole period is ‘the same’ , hence there will be no consumption at all, and no preference demonstrated. whereas if the ice is preferred sooner, than it will be consumed.
Ice-in-the-winter versus ice-in-the-snow is not keeping the good constant.
Ice-in-the-winter vs. ice-in-the-summer is keeping the good constant. The ice has not changed. You could swap the ice cubes from december with july and the consumer would not notice. You could switch winter and summer with summer, or talk about 2009 vs. 2010 or today vs. tommorrow or whatever. You can make this common sense case over and over again. Sandwich at 7AM vs. sanwich at 1PM is keeping the good constant. The conditions are changing as they always do. Again, that is what time does. Conditions necessarily change over time. And that is why Murphy argues people demand money w interest - it’s a rent on present liquidity given the uncertainty of future conditions.