Much of the benefits of modern life we enjoy and take for granted arises from scale economies in manufacturing — the larger the quantity manufactured by a firm, the lower the average cost. But there is a countervailing effect. Up to a certain size, the overheads of managing a firm goes down as the size increases. Beyond that point, the costs of managing go up with size. That limits how large firms can get. The market weeds out firms that grow too big because the inefficiencies show up in higher cost of production. That process is not unlike Darwinian natural selection in the biological world. There are creatures of all sorts of sizes but there are natural limits to how large even the largest can become. Become too large and you get sorted out.
Manufacturing has inherent scale economies and the efficient size of a firm is pretty large. Automobile manufacturing facilities are huge, for instance. But even there, there are more than one firm in the market. Moreover, no auto firm manufactures all its cars in one giant facility. There is decentralization.
That’s for firms which manufacture products. Firms that produce services have much lower limits because services don’t have the type of scale economies that manufacturing has. Beyond a certain point, the larger the service organization grows, the costlier is the service because the management overheads start to predominate the other costs.
The prototypical example of a very large service organization is the government of a country. The larger the country, the larger the government bureaucracy. At some point, the cost of the bureaucracy gets too large and services are inefficiently produced and distributed. Decentralization is the obvious strategy to deal with it. Break it up into smaller administrative units and not only do the management costs fall, but there is a possibility that competition among the units can lead to more efficient outcomes.
(As an aside, recall that Ma Bell was broken up and the Baby Bells competed amongst themselves and the US economy gained.)
* * *
Some mega-projects make sense. Suppose you were to build a rail network as the backbone of a national transportation plan. It makes sense to have a centralized authority that has the overall control of the project. It then subcontracts the various bits to private sector firms. Some firms can do the laying of the lines, some the building of railway stations, some with the electrical power network, some the signaling bits, etc. Centralized control but distributed actual implementation to specialized entities.
Public utility services are good candidates for mega-projects. Scale economies of course play a part in them. But more importantly, it solves a coordination problem — that of aggregating the demand of a very large number of users of the service and efficiently allocating the costs.
* * *
India has to supply the educational needs of a very large number. For up to the high school level, the estimated number is between 100 and 150 million. That’s roughly about half a million schools with an average capacity of 300 or so. Is it possible to have a mega-project which will create build, administer, and operate these schools? It sounds like a viable mega-project but it actually isn’t.
The cost of this project will be in the order of hundred billion dollars. Much of that will be swallowed by the huge bureaucracy that will be needed to administer the project. The informational requirements of that bureaucracy are impossible to meet because the product is not a homogeneous good. Education is highly differentiated and supplying it efficiently to a very heterogeneous market demand is exceedingly difficult because local demand is hard to estimate.
* * *
Public funding of primary and secondary education for those who cannot afford it is an absolute imperative. Primary and secondary education have public good characteristics. This has to be done in terms of an outright grant in most cases and in some cases, a bit of subsidy. For tertiary education, loans — not grants — are appropriate because tertiary education is largely a private good.
Public funds have been ear-marked for funding education. The question is how to use them most effectively. You could take the mega-project course and create a huge bureaucracy to manage it, build hundreds of thousands of schools, train millions of teachers, and more or less force people to enroll in those schools. You would do that if you had control of the billions of dollars of funds to do as you please. Handing out contracts for building schools to large contractors is a great way to make a bit on the side. Controlling the hiring of millions of teachers gives you a lot of power — and votes.
Alternatively, you could do a simpler thing. Instead of using the funds to build schools, you could fund the education of students by directly giving them the financial assistance that they need for them to pay for education. That’s called funding the student and not the schools. In other words, go with a universal voucher system.
Suppose for arguments sake that somehow it has been determined that primary education costs Rs 200 per month per student, and secondary education costs Rs 400 per month. Give everyone who does not have a primary education — without any discrimination — Rs 200 per month worth of vouchers. They then go and pay for the schooling to any school that does the job. Given so much money for primary education, schools will enroll students. Schools that do the job will succeed or else they will fail. Currently, government schools are given funding regardless of whether they deliver the goods or not.
This involves no large bureaucracy. Handing out vouchers — not money — cannot be as complex as building schools and all the rest of it. When the vouchers are redeemed by the schools, only then money changes hand. And the schools don’t get the vouchers unless they do the job. There has to be an independent regulatory authority which can oversee the performance of the schools to see that they are indeed functioning as advertised and that they are not simply fleecing the students.
In other words, this is a publicly funded but market delivered primary education solution.
The same goes for the secondary education except that only those who have finished primary education are eligible for secondary education support. Once again, give vouchers to all who qualify without discrimination and let firms build schools to do the job.
Tertiary education requires loans and not grants, as I said before. Give loans whoever requires it. If you get admission in an institution for tertiary education, you are eligible for a loan from a bank. Free entry into the tertiary education sector will ensure that bad firms will get weeded out. A bit of light-handed regulation would take care of any market failures in this regard.
* * *
Education cannot addressed the same way that you need to address the problem of transportation or energy. Mega-projects in education are unlikely to work in India considering that the Indian government is not known for its rectitude nor for efficiency.