Public and private goods: Part One

Both economic pedagogy and broad political discussion are based on what I’ve come to think of as anorectic understanding of public and private goods – which boils down to the idea that for things to go on well (let’s say in an economy) you need a mix of public and private goods with the former provided by government, the latter provided by markets.

I have a theory, or perhaps it’s better to call it a ‘framework’ or ‘paradigm’ in which all things of any complexity and value in life – whether economic, political, social, aesthetic or whatever – are the product of an ecology of public and private goods or to put it another way an ecology between collective and competitive activities. I think that’s a much more interesting way to think about things because it captures the potential multidimensionality and subtlety of the issues.

And though they don’t use the language I’ve used above, both Adam Smith and Friedrich Hayek painted a picture of the private activity of trading co-evolving with public goods.Thus, the public good aspects of markets (a marketplace, the ‘miraculous’ information system of market prices and liquidity) all co-evolve with the private trades that occur within the market. Governments usually turn up fairly late in the day to validate various collective arrangements and institutions that already exist or are coming into existence. The same ecology of public and private goods is necessary in information. These ideas were elaborated a little in this recent Lateral Economics paper (pdf).

Public and private goods are richly interdependent. Though he did not use the terminology we use here, one might restate Hayek’s argument by saying that it was about the quality of public goods. Hayek was arguing that the spontaneous order that arose through markets was a public good. We can be more specific and say that (given certain conditions) the prices that emerge in a free market are an emergent public good operating as a free, economy-wide information system available to all to use. A central planning apparatus could be described as a public good as well. Yet Hayek’s point was that it was of very poor quality because it could not take into account all the disparate information that is captured in a well-functioning price system. By contrast in the right circumstances, a market price system provides an incredibly efficient single quantitative indicator of the relative value of a given commodity (its price) considered as a function of its utility to each and every potential buyer and its cost of production to each and every potential seller.

In fact 1 the miracle of the price system is an example par excellence of a free rider opportunity. Price information between buyer and seller is privately known to them and indeed in thin markets can be tightly held as a secret between them often enforced by contractual obligations. However, in the kind of markets Hayek was considering, it was either impossible to conceal prices or the parties did not seek to. In either event the free riding opportunity was taken. The figure below illustrates the cumulative causation by which private and public goods grow together. Private goods when traded help build a market which generates the public goods of price discovery and liquidity which makes the market more useful for private transactions which further reinforces the emergent public pubic goods being generated by the market – with improved price discovery and deeper liquidity.

Public and private goods co-evolve in a market for goods and services

There is a similar ecology between public and private goods in the way non-price information circulates in markets. Information is encoded in standards and these standards are public goods. We discuss their significance below. Further as trades take place, non-price information that was not strictly necessary for transactions to be completed is acquired. Buyers discover aspects of the quality of what sellers are offering and/or vice versa. Like price information, this information is a potential public good. Thus for instance, if I need knee surgery, I can benefit from knowing which surgeons and hospitals have performed knee surgery best. But there may be no standard way in which such things are reported, or the standard may not surface information in a way that is useful to me.

Public and private goods co-evolve in a market for information
 Where price information is simple (involving a particular number) and often emerges in a market or can be easily engineered to emerge (as for instance it does on the stock exchange) non-price information may be complex and context dependent. In such circumstances it may be difficult to report meaningfully (the standards problem) and/or may not naturally circulate in the community. This frustrates the emergence of the public good from all the knowledge implicit in the private transactions.
  1. while the free rider problem bedevils economic activity in various respects[]
This entry was posted in Economics and public policy, Innovation, Philosophy, Political theory. Bookmark the permalink.
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
11 years ago

Hmmm, interesting that most of the real instances of marketplaces (e.g. stock market, or the local fresh fruit wholesale market, or even auction houses) are all privately owned… even though I agree that the information they produce is a public good. Strange too that you might have competing private firms (e.g. two different stock markets) attempting to outdo one another in delivering a public good (i.e. pricing information) which is mostly a non-rivalrous good. I say mostly non-rivalrous because early access to price data is more valuable to the first person who gets it, but I admit that’s nitpicking.

Thus for instance, if I need knee surgery, I can benefit from knowing which surgeons and hospitals have performed knee surgery best. But there may be no standard way in which such things are reported, or the standard may not surface information in a way that is useful to me.

Yeah, reputation systems keep looking like they should give brilliant results, but somehow never quite fit together because the technicalities are complex and people will generally actively seek ways to game them. If you just post a bad writeup about a knee surgeon, then you will probably end up in a legal battle, and even if it turns out you are perfectly right, you still lose time and money and sleep over it. Not much incentive to contribute to the public good.

Just after reading your article, I went to read up on the betting odds for the US election next week, and I got to thinking about that. So there are a great many things that could influence the outcome of an election, but the betting combines all of them into just one simple value — the odds.

Suppose you had some reasonably cut-and-dried statistic, like the number of cases each year where patients pick up an infection during their stay at a particular hospital (that’s just the first example off my head); and you trust the guy who collects and publishes the statistic. Then you set up a betting system where people can put up bets on future values, a bit like derivatives against whatever is your chosen statistic. My point being that betting on a statistic allows two things:
* the information is somewhat anonymous and non-specific so the lawyers can’t do much with it;
* and it allows many sources of information to be combined into a single bet.

This sort of gets around the worst of the “no standard way of reporting” because actually the details get stripped away and don’t get reported, but people’s gut judgement does get incorporated into the outcome. I know that you guys have tried something a bit like this with Kaggle, but that’s more along the lines of data analysis, whereas I’m thinking about a method of data gathering from diverse sources…

Yes, people could game the reputation system by secretly placing bets that favour themselves, but if someone guessed that you were doing that, they would bet against you and the more you pump it, the bigger the arbitrage you would be paying into for keeping that reputation in place. If you see what I mean.

11 years ago

I like your ecology analogy because the links are many and complex, though the public (govt) side does not evolve in the same fashion as the private.

“Thus for instance, if I need knee surgery, I can benefit from knowing which surgeons and hospitals have performed knee surgery best. But there may be no standard way in which such things are reported, or the standard may not surface information in a way that is useful to me.”

True, but some information, if not deceitful, is better than none and no alternative has emerged. I’ll bet you won’t get knee surgery without some asking around first. When mum got sick last year the first thing we did was to start asking around about doctors and double checking recommendations. Relatively easy for us, the poor old public patient has to take what they can get.

What do you think about drug advertising? Would that information make the public better or worse off?

Nick De Cusa
Nick De Cusa
11 years ago

The important point is to know if you are Cliff Richard or Judas Priest. Are you a public or a private good? A public good is meant to go for near-maximum usage. So it doesn’t charge for big profits. Rather it profiteers during peak-time, and gives people a free ride, or nearly so, during the other times. With a public good it ought to be financed out of the surplus. With a public good their ought to be no bankers, or other charlatans in the room when plans are being made.

Nick De Cusa
Nick De Cusa
11 years ago

So having said the above what on earth is the deal with the Conroy version of the broadband network? This is an undertaking that just cannot make up its mind if it is Debbie Reynolds or Ozzie Osborne.

Torrens Hume
11 years ago

This is a nice post. Nice blog — first visit.
If I understand it right, the idea seems that the issue of quality information is a problem of asymmetric information. It’s late and I am not thinking straight, but I wonder whether there is also a relationship with pooling and separating equilibria.

If the conditions are right for a separating equilibrium e.g. I can tell good doctors from bad doctors because they correctly signal their quality, then the markets for doctors (both good and bad) will become more liquid, if the outcome is a pooled equilibrium, then the market will be less liquid.

But since liquidity affects the price and demand for goods (in this case doctors), the next question is whether the general level of liquidity in the market place might also affect the initial conditions that determine whether the equilibrium is a separating or pooling one in the first place (i.e. the costs and benefits of signalling). Because of the potential interaction between liquidity and signalling, you might end up with multiple equilibria.

There have been a few posts on the issue of liquidity on other blogs lately.
On the complementary nature of liquidity and velocity see:
On the links between liquidity and the quantity of money and the public good nature of liquidity see:

Oh, @Tel , good point about the rivalry of emerging price information