10/28/2009
Lecture 15

Investment and technology---should we subsidize fertilizer?

Economists often talk about output Y as a measure of various inputs/factors:
  Y = F(L,K,T,A)  ->  L-labor,K-capital,T-land,A-technology

Last time: how much should ITNs be subsidized?
This lecture: how much should an agricultural factor like fertilizer be subsidized.

Malthus initially said there are fixed resources and growth will lead scarcity,
  which will lead to death.  Ten years later, he said he was wrong.  People have
  been saying this original Malthusian idea for generations
  (Tom Friedman, some guy in Rome, etc.)

Productivity increases are what make this idea not hold so far.

Food production per capita has increased since 60s in all of world, whereas
  in subsaharan africa, nothing has changed.
It seems to be pretty correlated w/ fertilizer use and increase in
  specialized crop growing.

In U.S. today, 2% of population feed more than entire population (we export food)

We probably need a larger study to see if increasing use of fertilizer leads to
  crop increases, or whether its seeds, etc.?

The Malawi example through a study by Britain showed that $74M worth of fertilizer
  subsidy resulted in $120M-$140M in extra crops.  Perhaps they can just tax this?
  - need a system that can handle taxing the farmers w/ more output
  - need a system where farmers can store extra earnings across the years
  - need to factor in cost of extra beaurocracy

Some work by Duflo et al. in Kenya shows a fertilizer randomized experiment
  where farmers put aside two acres of land and one was randomly given fertilizer.
  - result---in every single scenario, resulting reneue of the fertilizer plot
    was higher, even if you subtract market cost of fertilizer

Malawi isn't the only one subsidizing fertilizer.  Subsidizer costs were:
  - .75% of GDP in India in 1999-2000
  - 2% of govt. budget in Zambia in 2008

Arguments against subsidies
  - undercut market
  - could use money on other things, like fighting malaria
  - they are regressive because companies with economy of scale can afford more
    subsidy and see more output than smaller mom and pop shops.  You can regulate
    against this by onyl providing subsidies to small farms.

World bank thinks subsidies on their own cost too much.  They suggest instead (or
  in addition) to shift the supply curve of fertilizer by investing in infrastructure,
  distribution, and market model of fertilizer.  Also shift the demand curve by
  doing research + demonstrations, access to credit, improve market information systems,
  protect from volatility.

Different paper from Duflo, Kramer, Robinson
  - "chicago tradition"/"neoclassical tradition"---farmers are rational profit maximizers
    -> so if there are no externalities, subsidies/taxes will make farmers act
       non-optimally.
    -> subsidies regressive: the rich benefit more
    -> (govt.) public distribution is full of corruption/waste
    -> overuse of fertilizer can be environmentally damaging
  - but if chicago tradition is right, and fertilizer is obviously profitable, why
    aren't farmers using it?
    -> demonstration farms and real farms w/ randomized study show benefits of fertilizer
    -> farmers may not know about it? the authors show that fertilizer has been used
       for decades, and even if you show a farmer the increase in profits from fertilizer,
       the farms won't use it themselves the next year at the median.
    -> farmers are credit constrained?  fertilizer is divisible---even small ammounts
       result in some growth, so it's not some huge credit bump you need
  - so basically, farmers are irrational/don't make the optimal decision for themselves
    -> they procrastinate getting to the store, and underestimate their chance of doing
       something better in the future
    -> there's evidence for this in the psychology literature, and Nudge suggests
       that people just go with the status quo (in the case of pension plans, but
       maybe it applies to fertilizer use).

Randomized trial in the Duflo paper gives farmers option to order delivery of fertilizer
  for next season once they sell this season's crops.  This results on more use of
  fertilizer, but doesn't lead to increased use of fertilizer in future seasons.
  That said, subsidizing 50% at time of purchase vs. givng them chance to buy
  it for 100% ahead of time, so there are ways to increase fertilizer use without
  expensive subsidies.