Tuesday, February 18, 2014

Mieke Welvaert explains how Cuba pays for its oil imports and the measures it needs to take to make the system work

Mieke Welvaert explains how Cuba pays for its oil imports and the
measures it needs to take to make the system work
Posted in Opinion February 18, 2014 - 02:54pm
By Mieke Welvaert*

'Brain drain' is something New Zealand is inherently familiar with,
particularly when those 'brains' cost hundreds of thousands of dollars
to educate.

In Cuba, thousands of students (both foreign and local) undertake fully
government funded medical training.

Unlike New Zealand, the Cuban government intentionally generates an
oversupply of healthcare professionals, services which can then be
traded for commodities, oil and foreign currency.

This attempt to attain a comparative advantage in healthcare services
results in an over-allocation of resources to medical training relative
to domestic demand.

However, the nature of this comparative advantage is fragile, as it is
based on the ability of the Cuban government to retain much of the
incomes its citizens earn as doctors overseas.

Recent news stories have highlighted the growing trend for Cuban doctors
to defect to foreign countries. Rather than forego thousands in implicit
income taxes, these doctors are choosing to take advantage of richer
pastures. In this time inconsistency problem, the Cuban government
stands to lose millions.

Related Topics
Opinion Infometrics Joseph Stiglitz Mieke Welvaert Cuba dynamic
comparative advantage medical education
Under a recent programme called mais medicos (more doctors) 5,400 Cuban
doctors have been sent to fill gaps in the healthcare service in remote
corners of Brazil.

The mais medicos programme provides a 10,000 real (about NZ$5,000) per
month salary, in addition to meals and accommodation, to each doctor in
the three year programme. Of this 10,000 real, Cuban personnel receive
800 real per month in cash and 1,200 real is sent directly through the
Cuban government to family at home – a way to support the economy back
home and, perhaps more importantly, as means to encourage Cuban doctors
to return.

The Cuban government retains the remaining 8,000 real (about NZ$4,000),
effectively as an 80% income tax.

Initially in the form of aid, Cuba has practiced medical
internationalism since the 1960s. Going from one doctor per 1,393
citizens in 1970 to an unprecedented one doctor per 159 people in 2005,
Cuba produces enough doctors to have thousands to spare.

This apparent comparative advantage has enabled Cuba to export medical
services as a means to facilitate foreign policy and in exchange for
much needed resources – which, since the collapse of the Soviet Union
and the 1990s tightening of the US embargo, has largely been oil.

Cuban medical internationalism really kicked off in in 2003 when
Venezuela agreed to provide preferential trade for Cuban exports, to
partake in joint investment ventures and, not least, to provide 53,000
barrels of oil per day in exchange for a number of medical services,
provisions and training. Non-tourism service exports have been the
primary source of foreign currency in Cuba ever since.

The medical internationalisation scheme is an attempt at Stigliz's
"dynamic comparative advantage" – the idea that a country can create a
comparative advantage by simply mass producing a resource that another
country lacks. By producing healthcare services at a lower marginal
cost, Cuba has coordinated its exports with the needs of others.

New Zealand also exports doctors, albeit on a far smaller scale (and
with little recuperation of their foreign earned incomes) - so why don't
we just create this same competitive advantage and mass produce doctors
ourselves?

Such attempts at coordination have a corresponding distortionary effect
on a country's labour market, in terms of both the choices available to
people, and the choices they end up making.

Although Cuban doctors are encouraged through medical training with the
incentive of a higher income overseas, Cuban doctors in Cuba earn little
more than a taxi driver.

The fact that doctors are known to supplement their income by having
side jobs in the informal economy, indicate a preference for the
provision of services outside of healthcare.

The existence of this informal economy, signals that the direction of
funding specifically into healthcare professions is a misallocation of
resources.

Secondly, in New Zealand, a six year medical degree costs our government
upwards of $300,000 per person in fees alone. It is fair to assume that
the cost of producing doctors would reduce should economies of scale be
reached, but pumping resources into personnel for six years will come at
a cost. Despite being cheaper relative to other countries, putting
people through medical school en masse is an expensive exercise, one a
government is unlikely to do if there is no return.

Lastly, New Zealand has few ways to ensure doctors stay under the New
Zealand tax system and even if, hypothetically, we did copy Cuba's
methods, we would soon face the time inconsistency problem Cuba is now
battling with.

Cuba choses to solve this problem in a way New Zealand may find
unethical – by restricting doctor choice and implicitly holding families
as a form of security.

The Cuban government provides 'free' medical training and Cuban doctors
generate billions in tax revenue by working for the government while
overseas. Up until now, Cuba and Cuban doctors have largely cooperated
(there has been but a small 2% rate of defection thus far). However, as
with every time inconsistency problem there is pressure for parties to
renege on their agreement for their own personal benefit.

Although the Brazilian mais medicos initiative is expected to generate
US$250 million per year for the Cuban government, it also poses a threat
to their carefully crafted internationalisation system.

Unlike previous 'doctors-for-oil' agreements, which were unique to the
Cuban government, the mais medicos programme is open to doctors from
more countries than just Cuba. Under the same contract as their
Brazilian, Portuguese and Spanish counterparts, the difference between
what Cuban doctors earn and what they receive is blatantly obvious. As a
result, Cuban doctors are threatening to leave.

If it were easy to defect to another country, there would be a hefty
incentive for school leavers to train for free in Cuba and to spend
their entire practising career elsewhere. Should doctors defect to the
likes of the United States, they could see their monthly incomes
increase from $15-$30 dollars (at home) to upwards of $13,000. To add
further pressure to Cuba's internationalisation system, in 2006, the
Unites States initiated a programme to fast track asylum for Cuban
healthcare professionals working in a third country.

However, in a society where one doctor's education is paid for by
another's taxes, it is important to realise that without the tax revenue
from 'medical internationalistas' there is no funding for medical
education (among other government funded goods and services).

Cuba provides an interesting case study in how government policies can
help individuals coordinate in their choice of job and educational
attainment.

However, it also shows some of the costs of trying to generate a
"dynamic comparative advantage": the misallocation of human capital as
people are pushed into specific areas, the restrictions to freedom
required to solve the time inconsistency issue between government and
the service providers, and ultimately the risk of being exposed to an
industry 'picked' by a government.

-------------------------------------------------------------------

Mieke Welvaert is a data analyst at Infometrics.

Source: Mieke Welvaert explains how Cuba pays for its oil imports and
the measures it needs to take to make the system work | interest.co.nz -
http://www.interest.co.nz/opinion/68547/mieke-welvaert-explains-how-cuba-pays-its-oil-imports-and-measures-it-needs-take-make-

No comments:

Post a Comment