Bad Ideas for Ending Global Poverty (Part I)

A specter is haunting the world — sector-specific government support for the information and communications technology (ICT) industry.

Take the case of Malaysia. This Southeast Asian country has launched perhaps the most ambitious set of incentives in the developing world with its Multimedia Super Corridor (MSC).

The MSC began in the mid-1990s with promises of $10 billion worth of public infrastructure investment, a government venture capital fund and significant tax breaks, amongst other incentives, to companies that would set up in the corridor.

India, for its part, has a raft of subsidies and tax breaks available for those who set up in the country’s growing number of IT parks. Numerous other countries — as diverse as Russia, St Lucia and Rwanda — have launched programs to attract ICT industries.

Proponents of government subsidies argue that ICT is a high-productivity sector that can kick-start economic growth in developing countries. At the same time, evidence that ICT firms “cluster” — converging on areas like Bangalore or Silicon Valley — suggests to some that there is a place for subsidies to attract a few initial firms with the hope that a lot more will then follow.

Both arguments, however, turn out to be a little threadbare on further analysis. The high productivity doesn’t benefit local producers, and evidence that subsidies alone can foster a self-sustaining ICT cluster is very weak.

It is a commonplace that recent productivity gains in the United States and elsewhere have been driven in some considerable part by the fact that the ICT industry has been making faster and more powerful hardware and software at the same price.

And the sector is a large one in a number of developing countries — accounting for as much as 21% of Malaysia’s GDP. But this doesn’t necessarily mean that there has been a considerable impact on the quality of life of Malaysians as a result.

That’s because the country exports most of its ICT industry output, and it is consumers, not producers, which are gaining from productivity improvements.

These consumers, largely resident in wealthy countries including the United States, are getting better machines for the same money — which translates into the fact that Malaysian firms are being forced to produce more computing power, or better televisions, for the same price.

Whether it should be the first priority of the Malaysian premier to foster policies to benefit the U.S. consumer is, surely, a question that might spark debate in Kuala Lumpur.

And this is a powerful argument against developing country governments subsidizing ICT industry investment. There is no more reason to support the commoditized ICT sector than the agriculture sector on the grounds of different returns to productivity.

There might still be grounds for government support if a small amount of subsidies would kick-start a substantial, self-sustaining industry.

And there are occasions where such an approach appears to have worked. For example, East Asia’s hardware industry did receive tax breaks, subsidies and research support and has grown very large.

Nonetheless, this is a high-risk strategy because subsidies are a very small part of the picture that firms look at when deciding where to invest, and such subsidies may become an economically inefficient, expensive, growing and ongoing fiscal burden.

Malaysia’s multimedia supercorridor is a particularly powerful case study here. The country launched the MSC to expand out of ICT hardware production (where it was already a substantial player) into software and convergence technologies. The $10 billion-plus investment by the government was matched by just $475 million of private investment and 7,300 jobs (that works out at more than $1 million per job) up until 2000.

Reasons that companies cited for not moving to the corridor included concerns about government monitoring of Internet traffic, capital controls, red tape, slow visa approval, weak intellectual property rights and the absence of an appropriate skills base.

At the same time, India’s incredibly successful software boom — forecast to be worth $60 billion in exports in 2008 — has been linked to tax breaks and subsidies to the industry by government.

Indeed, as early as 1986, India’s software promotion policy included tax holidays, tariff breaks and export subsidies, and by 1990 software technology parks were being established. By 1999, such parks accounted for 68% of the country’s IT exports.

But most of the tax breaks, subsidies and parks came after the industry had already started its growth, not at the birth of the cluster.