Subsidiarity is one of the guiding principles that glues together the sometimes reluctant nations of the European Union (EU). In essence, subsidiarity holds that if something can be achieved at a lower level of government rather than a higher one, then it should be done at that lower level. It has proved a powerful tool for those resisting centralist tendencies within the EU institutions. It is also the principle that determines that while the United States fights wars as a nation, its 50 states collect sales taxes independently and create their own legislation.

The distribution of startup capital to companies in biotech and other high-technology areas has become a victim of subsidiarity and a drain on the public purse to boot. As venture capitalists move away from investment in highest-risk, early-stage ventures, a growing number of national and regional government bodies throughout Europe, often in partnership with local banks, have stepped into the seed-funding breach. They establish small funds that can invest or co-invest in ventures, including biotechnology. Almost without exception, the funds have a remit to invest exclusively in local firms. There is, in principle, nothing wrong with investing locally: it's pretty much what most venture capital companies do all the time. The problem arises when the investor only takes a local view of the investment.

For instance, a major sponsor of November's Bio-Europe partnering conference in Cologne was the Stadtsparkasse Köln, one of Germany's largest commercial and savings banks, which is based in Cologne. In a plenary session and subsequent press conference, the bank's CEO, Gustav Adolf Schröder, made great play of the bank's long-standing support for biotechnology in the Cologne region. The reality is somewhat different. The bank has indeed put money into 12 life science companies in or around Cologne in the past few years. However, it has invested only €8 ($10.4) million in total, a cheese-paringly hesitant investment strategy in the context of biotechnology's long haul. Two of the 12 companies are in reasonable shape but the rest, apparently, are a little shaky and may soon be saying auf Wiedersehen.

A Schweinsgalopp into European biotech by a regional bank is not unique to Cologne or the Stadtsparkasse. According to a background document prepared for a workshop on biotechnology financing organized by The Netherlands as part of their presidency of the European Union (http://www.minez.nl/), the average European biotech startup under two years old received only €1.6 ($2.1) million in 2002. The equivalent figure for US companies was just under €7 ($9.1) million. Destined to scrabble around for cash from the very start, European biotechnology companies never flourish, it appears. The same report indicates that investors have pretty much thrown in the towel by the time a European company is over 5 years old. Only 15% of older European companies were backed during the two years 2002–2003 and then only at around €10 ($13) million a time. In the United States, the proportion of older companies getting backing was twice as high and the amounts each company received were three to four times higher.

Whatever the reasons for underinvestment, subsidiarity is not fixing the problem. Small local funds allocate small amounts to small companies. More dangerously, they are often investing in the wrong companies because they take a regional view of investment. The classic example has been the BioRegio initiative in Germany, where federal and state money was used to match venture capital investments in biotechnology firms. This resulted in the creation of hundreds of new companies, only one-third of which (according to Ernst & Young) now have enough cash to last more than 12 months. Many are now the reluctant subjects of enforced 'merger' discussions, causing key biotech industry representatives from across Germany to go back to the drawing board; as of last month, the plan is now to reunite Germany's stuttering regional initiatives.

In 2000, well after BioRegio had started, the Dutch government also joined in the party, putting forward €45 ($58.7) million into a scheme called 'BioPartner' with the express goal of creating 75 new companies. The goal, as is the nature of these things, was achieved. However, there are now around 75 new companies in the Netherlands looking for more cash—and it is clear that no more cash is going to come from the Dutch government.

There is nothing intrinsically wrong with government schemes that attempt to address biotechnology-funding gaps in which private investors are not operating. In fact, because EU competition rules restrict the amount and type of help that governments can direct to local companies, the scope for regional support may be greater outside the EU, especially perhaps in US states outside the main clusters in California and New England, or in countries that are isolated from international finance, such as Australia or New Zealand.

But make no mistake, these regional initiatives will be doomed to failure if they continue to combine implicit quotas of investments and a regional remit. The calculation is quite simple: in any given region, only 5–10% of the academic spinouts are likely to be competitive when assessed against international norms. Yet regional schemes, using only regional benchmarks, often fund up to 50% of startups. This might look good for politicians. But it is dooming many companies to an early grave and wasting taxpayers' money. Eighty to ninety percent of these companies are never going to compete at the financing round when the 'soft money' dries up.

The solution is not to do away with regional funds. That would be to lose forever the deserving 5–10% of projects that might make it commercially. But regional funds must get away from the company-counting mindset. And they must be managed by people who can take a detached, international view of the regional scene.