Export Promotion Organizations XPOs Are they Useful

1
Export Promotion Organizations XPOs Are they
Useful?

• By Andrew Singer, May 2006.

2
How the XPO model came about

• From 1950 to 1962, the developing country share
  of world imports crashed from 32 to 21. This
  was seen as a crisis, and required a response.
• Major UNCTAD meeting in 1964. Recommended a
  permanent center in each LDC for trade
  information and market research.
• The concept of a centralized XPO or TPO was
  born.

3
The Model Quickly became Universal

• Most developing countries set up XPOs most in
  the 1960s and 1970s.
• UNCTAD established the International Trade Centre
  in 1964. Its main job was to help set up these
  XPOs, and get them started.
• Most were set up within a ministry, or as a
  public agency under a ministry. Nearly all
  provided services free of charge. Nearly all
  were staffed entirely from the civil service.
• This is the model that still persists. I
  estimate there are still over 70 XPOs in
  developing countries.

4
The World Bank Research Project of 1991

• The first major research project on how this
  model had worked in practice.
• Undertaken by Prof Don Keesing Stanford World
  Bank and myself.
• The major conclusion that the XPO model, that
  had dominated aid support for developing country
  exports for over three decades, had not worked
  well.
• We searched for successful examples. Experience
  had nearly always been negative.

5
The World Bank Research Project of 1991

• My work since 1991 has served to confirm these
  conclusions.
• Even the worst of the XPOs refuse to die. Aid
  agencies refuse to let them die.
• Giving aid to support exports is just too
  attractive. If an agency exists to promote
  exports, then it is almost bound to be supported,
  no matter how ineffective.

6
Why the Model did not Work

• What happens in countries where exports have
  expanded successfully.
• Services for exporters are provided by a
  plurality of mainly private service providers,
  many of which are highly specialized, and provide
  highly technical services.
• What seems to work is a marketplace, supplied by
  lots of specialists.
• Where an XPO exists, its impact is accepted as
  being limited.

7
Why the Model did not Work

• The initial expectation in 1964 that one single
  service provider would provide the missing
  ingredient for export take-off that expectation
  was, in my view, fundamentally wrong.
• Placing this single service provider in the
  public sector, and staffing it from the civil
  service that just made a wrong model even
  worse.
• Concentrating on market entry services, rather
  than dealing with supply constraints, that was
  also, in hindsight, a big mistake.

8
The Concentration on Market Entry

• In recent years, that mistake has been broadly
  accepted.
• As OECD markets have opened up to developing
  country exports, these exports have not achieved
  a broad take-off. eg. AGOA, EBA
• It is now understood that the essential reasons
  are on the supply side.
• Market entry, where XPOs have so long have
  concentrated their efforts, was never the issue.
• Understandable. Market entry services need much
  less technical specialization and are
  politically much more attractive to host
  governments.
• Too often, in my experience, concentration on the
  XPO has been used as window-dressing, to take
  the spotlight off difficult policy-related supply
  constraints.

9
Given the Experience Can they be Useful?

• Yes, so long as it is accepted that 90 of the
  effort and aid needs to be put into dealing
  with supply constraints, where XPOs do not, in
  my view, have a useful role.
• Another condition to be useful, front-line
  staff need to have direct export experience.
  Civil servants generally wont do. Second
  experienced export professionals from the private
  sector.
• Concentrate on firms that are new to exporting.
• Provide them with seconded advisors.
• Provide training in broad cross-sector skills
  eg. how to find an agent, how to deal with the
  paperwork, etc.
• Any subsidies eg. trade fairs should be
  strictly time-bound.
• Service users should pay a significant and
  increasing proportion of the full costs.