Developing countries are often advised (or instructed) to undertake reforms recommended by "experts" who are called "technocrats" and are often backed by the IMF. Opposition to the reforms they propose is usually dismissed as "populist." Countries that fail to undertake these reforms are dismissed as craven or lacking political will, and soon suffer the consequences: higher interest rates when borrowing abroad.
But look closely at some of these "technocratic" proposals: many are more often based on ideology than economic science. Technocrats can of course reliably make an electricity plant work better. The goal is simple: to produce electricity at as low a price as possible. This is mostly a matter of engineering, not politics. Economic policies are usually not technocratic in this sense. They involve tradeoffs: some may lead to higher inflation but lower unemployment; some help investors, others workers.
Economists call policies where no one can be made better off without making someone else worse off Pareto efficient. If a single policy is better than all others for everyone, ie, has no Pareto-efficient alternatives, it is said to be Pareto dominant. If choices among policies were purely Paretian, ie, if no one was made worse off by choosing one policy, as against another, the choices involved would indeed be purely "technical."
But in reality, few policy choices are Paretian. Instead, some policies are better for some groups, but worse for others. Different policies benefit and hurt different groups.
In East Asia, for example, IMF bailouts helped international lenders, but hit workers and domestic firms hard. Different policies might have imposed more risk on international lenders, and less on workers and domestic firms. Deciding which policy to choose involves choices among values, not just technical questions about which policy is in some morally uncontroversial sense "better." These value choices are political choices, which cannot be left to technocrats.
Of course, there is scope for technical analysis even when political choices are at the crux of the decision. Technocrats can sometimes help avoid Pareto-inferior policies, that is, policies that make everyone worse off. Sometimes there are policies that can promote both growth and equality, and the job of good economists is to search for them. The problem is that many policies advanced by technocrats as if they were Pareto efficient are in fact flawed and make many people -- sometimes entire countries -- worse off.
Look at the litany of technocratically inspired examples of privatization and deregulation in the 1990s. Banking "reform," for example, frequently soon required government bailouts, leaving a few people much richer, but the country much poorer. These failures suggest that we should have less confidence in the supposed professional skills of technocrats -- or at least less confidence than they have in themselves.
But there is also a more fundamental point here. Democratic processes are likely to be more sensitive to the real consequences of policies, to the real tradeoffs involved.
Of course, some criticisms of technocratic remedies may be populist posturing, but sometimes they contain insights that ivory-towered (and usually US-trained) technocrats miss. Consider the case of Mexico, where a proposal to raise revenue by taxing food and medicines consumed by the poor was, unsurprisingly, rejected by a democratically accountable legislature.