In most countries, the award of a Nobel prize is a cause of much self-congratulation. But not in Germany. When Theodor Hänsch won a share of this year's physics Nobel for his work on laser-based precision spectroscopy, many commentators saw the award as an exception to the rule of underperformance in German science.

Such introspection is all too characteristic of modern Germany. The truth is that German scientific prowess and inventiveness can still compete with the best in the world in many spheres. But the science system suffers from a number of structural weaknesses. The new government, almost certainly a ‘grand coalition’ of the two large political parties, the Social Democrats (SPD) and the Christian Democrats (CDU/CSU), should move quickly to address them. And the science ministry, which is likely to be led by Annette Schavan, a former minister for education in the state of Baden-Württemberg, should focus on areas where intervention is really needed.

It should start by revisiting Germany's restrictive rules and regulations in stem-cell research and plant biotechnology, which make it hard for researchers in Germany to be competitive. Former chancellor Gerhard Schröder had promised to open discussions on regulations that effectively prevent research using embryonic stem cells. Schavan said she wouldn't. This is one promise she should break.

The new government should work to create conditions that will keep women in research.

Despite its long tradition in science, Germany still lacks research universities that can compete with the best in the United States and Britain. The federal education system, which distributes strong universities all over Germany, has served the country well. But to compete internationally, Germany needs to target a few institutions and support them to the hilt. The previous government launched a €1.9-billion (US$2.3-billion) programme to strengthen the research profile of the best German universities. The new government will need to push this programme hard if it is to overcome state rivalries and allow the strongest research universities to grow stronger.

Across the university system as a whole, reforms are needed to ensure that appointments are made on merit. At present, political considerations and personal influence are often more important than a candidate's scientific strengths. If institute and department heads, and not ministry officials, were solely responsible for making appointments, this situation would change. Schavan says she wants the federal government to retreat from its joint responsibility for universities, and this could prove to be beneficial provided that the state (Länder) governments grant increased autonomy to the universities afterwards, as many of them have pledged to do.

Given a longer leash, universities could work harder to attract the best scientists from abroad. Germany has failed to make the most of the scientific talent in eastern Europe, despite its geographical proximity. While US universities rolled out the red carpet to immigrant scientists from that region, most German universities have not been flexible enough to create new positions, or to offer competitive salaries and conditions. The new government must help to put that right.

It should also take steps to make research careers more attractive for the declining number of science students in Germany itself, and should create conditions that will keep women in research. It can start by nurturing merit-based appointments and providing incentives for employers to provide decent child care (see Nature 437, 296; 2005). Schavan, who calls herself a ‘conservative feminist’, and Angela Merkel, the chancellor-designate and a former physicist, need to find ways to help more women achieve in science what they have done in politics.

Finally, money counts. Scientists will expect both coalition partners to stick to their pre-election commitments to gradually increase public science and education budgets. This also has a European dimension: it is unlikely, without Germany's active support, that the European Union's next Framework research programme will develop as it should.