Up until now, agriculture has been the domain of professional agriculturalists with a narrow focus on increasing productivity.

Merely having people leave the land for big city slums does not improve agricultural yields, as linca wrote:

The majority of peasants in the third world are stuck with agricultural methods that went out of fashion in the High Middle Ages... And the productivity that comes with it : whereas these tools only allow a worker to cultivate about 1 hectare,  or 2.5 acres by himself, and yield thus 1 ton of grain from his work, a 1900 European or American peasant, using animal traction and basic industrially produced tools, could cultivate ten times the area, and get about 10 tons of grain for his work, and thus was already 10 times more productive ...

As the use of hand/animal labor was not able to raise the total yield of the land but concentrated the improvement - 10 peasants producing 10 tons to 1 peasant producing 10 tons.  

Second, the nine who left the land didn't evaporate!  They moved to major cities where they compete for lower paid, unskilled, jobs implying supply-side driven wage reduction.  In turn, this indicates the market price of agricultural goods produced in that country will lower as the Purchasing Power of those jobs is divided among more people.  

Third, the migration of people from rural to urban areas enfeebles the rural economic structure as the number of consumers lessens.

This trend is countered by the introduction of (however minimal) Public Health measures by various organizations increasing the rural population.  From a humanitarian stance lowering the child death rate is a good thing.  From a systems stance, not so much.  A general trend in Africa, for example, is that women and children stay in the agricultural areas while the men go to the city to try and earn some money, returning intermittently.  Since, as Asinus Asinum Fricat observes, women are not empowered this intermittent return implies Yet Another Kid.  

The last thing anybody or the planet needs.

The given examples are from Africa.  The same general system is in place in any country conforming to the same general pattern, e.g., rural Appalachia in the US., albeit with different dongles; the dance is the same, the steps are unique to the economic and technological situation of the country.

While disheartening I submit this opens the possibility that solving the Third World food crises would go a long way to solving the problems of rural poverty in the First World as well.  If this is shown to be correct it would give 'cover' for First World politicians to effectively act to change policy.  

The key, as I've already said, is to broaden the pool of experts influencing First World agricultural policy making.  As an added benefit it would broaden the pressure group(s) beyond those directly benefiting from the current policies.  Those people are powerfully connected to and in the political structures of a country.  It has been said, elsewhere, the largest beneficiary of EU agricultural subsidies in the UK is the Royal Family.  I'll bet a dollar to a bucket of cow manure investigation would uncover similar situations in the other EU countries.  It will be hard to overcome their influence.  If it isn't, nothing will happen.

Before ending this ramble let me say: eliminating ag subsidies would be a disaster, effectively gutting agriculture in the US and EU.  Whether or no in some different reality ag subsidies aren't required in this reality -- you know, the one we live in? -- it can't be done.  

What needs to be done is a wide-ranging, embracing, policy change to move the global agriculture system from here to there.  Where "there" is a sustainable agricultural system in vibrant, economically viable and diverse, rural areas.


She believed in nothing; only her skepticism kept her from being an atheist. -- Jean-Paul Sartre