Some notes on public goods

taxonomy of goods

I was going on about the renewed importance of public goods to the Review Panel on the Innovation System and so they asked me and another economists on the panel to do a bit of a write up for them. For various logistical reasons, the ultimate document was run up by me the night before the next meeting. I’ve reproduced it below the fold with a few nips and tucks mainly to remove typos etc. It’s not rocket science, and readers of my stuff on this blog won’t find anything much that’s terribly new, but nevertheless, for the record, it’s below.

The following notes have been made rather hurriedly to flesh out some comments made in the last meeting. They are intended as an introduction to discussion, rather than as anything definitive.

What is a public good?

The technical definition of a public good is something that is non-rival in production and non-excludable in consumption.

Non-rivalry in production means that supply to one enables supply to all. Examples include clean air, defence of the realm and a TV broadcast. The first of these two things are non-excludable also meaning that if you provide them for one, you cant prevent others accessing them and this makes them pure public goods. TV broadcasts are different because you can encrypt them and then get people to pay for the decryption.

The general theory of public goods says that they are under-produced unless provided collectively. Thats because non-excludability prevents the producer from capturing the full benefits of what they produce. This is true of defence which is often regarded as the classic public good. But there is actually a class of public goods that are generally provided privately, though they might be better referred to as social assets rather than public goods.

This is the web of social understandings comprising social mores and most particularly language. Language is a public good, but it is not produced by governments or by collectives acting as collectives. It is produced by individuals who are already part of a community (the community that speaks the language).

Economists have typically concerned themselves much less with the former kind of public good than the latter, (the exceptions include Adam Smith who wrote a treatise on the formation of language and a bestseller on culture, socialisation and social mores The Theory of Moral Sentiments which incidentally sold much better than the Wealth of Nations during Smiths lifetime. And Friedrich Hayek who was also concerned with the way in which such public goods emerged spontaneously from the private actions of many people seeking mainly to advantage themselves.)

There may already be a jargon term in use, but thinking about this, I make the distinction between mechanical public goods and ecological or organic public goods. The former generally require some deliberate collective funding to be produced in the appropriate quantity, while the latter seem to do OK without this because their creation is cognate with their use.

Mechanical, Collectively Funded Public Goods

Ecological, Privately Cultivated Public Goods


Social expectations those that make society function

Lighthouse, Traffic signs (local public goods)


Law as a sovereign command (Statutes, regulation)

Law as social custom (common law), industry standards.

e-tax financial software funded and freely distributed by the ATO

GnuCash free, open source financial software.

Public goods and the public sector

Even where goods are excludable, the fact of non-rivalry creates a case for governments to get involved. That is because if governments fund the production of non-rival goods, they can then fund them through the tax system and make them available to the community at zero marginal cost. In principle this will often be more efficient than production by private agents who will charge too much for their distribution. One might justify the ABC on these grounds, though it is generally justified on the grounds of promoting natural culture, and improving access and information provision to rural citizens.

In any event, this led me to propose in an article published in the late 1990s that where governments produce goods or services, the presumption should be that the public should have access to them at their full cost of production. There is a public interest argument here loosely analogous to freedom of information. And theres an economic efficiency argument. If goods, services or assets are in existence or can be produced at a cost people want to buy them from a government agency, then the government agency should have good reasons for refusing them.

Thus (I argued), for at least as long as the public owned Qantas and the Commonwealth Bank, other businesses like new airlines and banks should have had access to their assets (at full commercial cost). In fact the Commonwealth Bank and Qantas acted restrictively and, given that this is very likely to have led to sub-optimal economic outcomes, it was silly of the public owner not to have instructed those businesses to provide others with access.

However something important has been going on since the rise of the internet. It is this.

While our reform of our economic institutions has focused on removing collectivism in various forms generally rightly so new technologies, most particularly the rise of the internet have given rise to a plethora of new public goods. Generally these have been ecological public goods built up by the accretion of small private acts.

Thus the internet has spawned open source software Linux, Firefox, Open Office, Apache. These have been produced by programmers in a way analogous to the development of language. Everyone has contributed to everyone elses code because it was in the interests of those who participated to do so, and the internet made such far flung collaboration possible. The code has now built up into powerful programs which are becoming more powerful by the day. Even without code, it turns out that social cooperation can produce major public informational goods like Wikipedia.

The new means of production are bringing into existence goods and services which would otherwise not exist like (a service by which local people with the same interests can discover each other and meet) and replacing private provision in some areas for instance encyclopaedias and proprietary software.

These issues are illustrated in the following table.

Private Goods

Mechanical, Collectively Funded Public Goods

Ecological, Privately Cultivated Public Goods

Microsoft Money, MYOB, Quicken

e-tax financial software funded and freely distributed by the ATO

GnuCash free, open source financial software.

Ironically, although government agencies are better placed in theory to develop such products which rely on a high level of collegiality and are assisted by some sense of altruism (though this should not be overstated) in practice governments have helped very little. Though they dont produce for profit, and so dont have pecuniary incentives not to co-operate, in fact governments are very poor at the kind of loose, exploratory, collegiality and collaboration that characterises open source production.

Governments could well make things worse if they tried to involve themselves too heavily in open source production, but there are a range of things that they can and should consider doing to help foster the development of open source ways of doing things. And the new forms of collaborative production so called web 2.0 offer incredibly exciting ways of improving governments capacity to consult the electorate and to involve it in its deliberations.

The internet is also bringing together production which is driven by both profit and higher i.e. more altruistic motives. Because knowledge goods can be scaled into global goods at near zero marginal cost via the internet, it is becoming possible to create massive amounts of value at low cost. The entrepreneur then needs only to harvest some tiny fraction of the value of the good or service they have produced to cover its costs very profitably indeed, while leaving the community massively better off.

This is essentially Googles business model seek value for customers on the assumption that if a lot of value is created, the creator will be able to turn some small share of it into a tidy profit. As Google CEO Eric Schmidt puts it High growth solves virtually all problems. Googles value to users is massively in excess of the value it harvests from ads. (Just ask yourself how often you went to Google without clicking on a Google ad link or more precisely without buying anything having clicked through.) Google is doing the same with Google News, G-mail, Google Maps, Google Desktop and virtually all its offerings with one hundred more in development!

Further it is clear that another public good is of immense importance and that is the existing stock of knowledge. Technological innovation has often gone hand in hand with new social conventions of openness. In the 1960s and 70s, Japanese car companies extended collegiality and collective problem solving to their employees and their suppliers. The result was Just In Time or the Toyota production system a huge increase in productivity now imitated around the world.

More recently Silicon Valley appears to have been built on the public good of the knowledge of the region, often freely shared between firms in ways that didnt happen in other, often better equipped regions. Here are some extracts from Regional Advantage: Culture and Competition in Silicon Valley and Route 128 by AnnaLee Saxenian (1994) in which she attributes the success of Silicon Valley over Route 128 (around MIT in Boston) to this new kind of openness. MIT had better human capital to innovate in IT, but most big things happened in Silicon Valley.

A Fortune magazine writer who visited Silicon Valley in the early 1970s described the ‘technological community” he found there in simi¬lar terms: “A surprising degree of cooperation among companies almost Japanese in its closeness, has added further impetus to Santa Clara’s ascendancy . . .

Information exchange continued on the job. Competitors consulted one another on technical matters with a frequency unheard of in other areas of the country. According to one executive: ‘I have people call me quite frequently and say, ‘Hey. did you ever run into this one?’ and you say ‘Yeah. seven or eight years ago. Why don’t you try this, that or the other thing?’ We all get calls like that. The president of the Western Electronics Manufacturers Association (WEMA) compared the openness of Silicon Valley to the East: “Easterners tell me that people there don’t talk to their competitors. Here they will not only sit down with you but they will share the problems and experiences they have had. According to an experienced semiconductor executive: This is a culture in which people talk to their competitors. If I had a problem in a certain area, I had no hesitation to call another CEO and ask about the problemeven if I didn’t know him. It was overwhelmingly likely that he’d answer (my question).

These issues are now reverberating around the question of the science commons and the extent to which existing IP regimes are helpful or inimical to innovation.

Issues arising

There are lots of issues arising from these considerations. Here is a list which makes no pretensions to comprehensiveness and which is in no particular order:

1. By what means might governments facilitate the development of open source means of production? How can they get out of the way for instance by minimising the exposure of open source projects to unduly adverse legal risks through IP and other legal liabilities (without undermining legitimate incentives for others).

2. Can they assist and if so how? Could they fund prizes, purchase IP to prevent legal obstacles or seed open source projects, provide funding to projects, donate the code they fund to open source communities or use purchasing to assist open source projects?

3. How can openness and collegiality of the infrastructural and informational knowhow behind so much innovation be maximised? And are there things that governments do that get in the way of this?

4. Can governments become more open in their deliberations, to maximise the extent to which those outside government can provide assistance in solving problems. Can they use web 2.0 style collaboration to improve the quality of their deliberations and/or reduce its cost?

5. Governments are not good at using the information that is distributed through society Hayeks argument for markets. But markets are often not good at handling the public good aspects of information and indeed often suppress information for instance on product quality. How can we optimise arrangements for the generation and dissemination of information to guide both consumers and producers?

6. Can the role of philanthropy be optimised to involve it more in the production of new public goods?

Ive posted on these issues at Club Troppo including outlining how the ABC could transition to becoming substantially more of a web 2.0 style operation without deviating from its central mission.

1. Two economic paradoxes of our time: Part One – the paradoxes
2. Two economic paradoxes of our time: Part two – what to do with a sketch of an example using the ABC
3. Doing well by doing good

This entry was posted in Economics and public policy, IT and Internet, Media, Science. Bookmark the permalink.
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
13 years ago

Non-excludability means you can’t price, non-rivalry as well means you wouldn’t want to even if you could. But excludability with non-rivalry does not imply a case for government. It might. Club goods can also be delivered by private interests as, well, a club.

You pay a fixed fee for access to cover the costs of the club and then have unlimited access at zero price. The ABC could be run along these lines – a $100 fee per year then all the ABC media you like.

Club goods can nicely fill gaps in the supply of public goods – e.g. private nature reserves owned by a club – that have members who have conservation interests stronger than the average tax payer who meets general community needs by voting for a party which reflects his/her average conservation interest.

13 years ago

Some quick suggestions which you have probably already thought of, since they are all obvious (for all of which none of them, afaik, have actually been done):

1 At least cut back copyright extensions to something reasonable, like 50 years capped, for example, or the author’s lifetime, or both.

2 Making all government-funded code, besides security-sensitive stuff, open-source seems like a no-brainer.

3 An easy answer would be to have more 2020-type things. I would prefer to have summer-camp style scholarship programs for bright and/or dedicated but ‘disadvantaged’ young people to spend eg 4 or 6 weeks a summer on a physics or engineering or chemistry or maths ‘camp’ with eg second and third year uni students doing most of the facilitating/teaching – this would increase breadth and depth of networks and radically increase penetration.

4 Yes but have you read Yes Minister? Yes, absolutely, and organisations like ASIC need to do this with urgency. Also it is tempting to say that draft legislation should be on the web for at least (eg) three weeks prior to parliamentary debate so that the community can comment. The downside is that in the cases of most legislation, ‘stakeholders’ already comment and non-‘stakeholders’ are often rarely able to understand what the legislation in question achieves – maybe this is where the focus should be: maybe the EM should be available on the web in fixed-wiki form so that the original text can’t change but people can insert comments wherever they like?

That is just in relation to legislation which is not really your point – but in the case of legislation this is already broadly within the current processes – it would be much more complicated to insert web-2.0 into actual departments. OTOH, much more could be done within departments, for a start.

5 Google has already done this and facebook is completing it. There is room for work such as your ideas on standards, but I would focus on improving the ability of the general public to understand the information already available to them – ie education.

6 As opposed to the public goods it already focuses on? I assume you mean funding the development of software, etc? Many philanthropic institutions will do this indirectly, eg through university funding. Maybe research funded to at least 50% by tax-exempt entities should be compulsorily open-source and not able to be fully patented/copyrighted, or perhaps compulsorily licenced for non-profit use? This risks conflicting with other goals in this sector but perhaps could be managed.

13 years ago

Might be of some interest to you here Nicholas,23599,23668471-23109,00.html

Mike Pepperday
Mike Pepperday
13 years ago

I once read that the most successful company in the world was the Silicon Valley Company – and it didn’t have an office or board or a letterhead.

I suppose this success was partly due to those CEOs talking to each other.

But why doesn’t it turn into Smith’s conspiracy against the public? When airlines (and they are just the latest) do it they are fined mind-boggling amounts. What, precisely, is the distinction?

Mike Pepperday
Mike Pepperday
13 years ago

Yes. If you talk to your competitors about how to lower costs you will be more competitive both toward other competitors and against completely different products vying for customers’ dollars.

So the essence of it for the regulator would be that you’re allowed to talk about increasing competition but not allowed to talk about decreasing competition.

Makes sense. Thanks.