IMF/World Bank civil society forum

Thursday 10 October 2013

Panel:
•    Chris Lane, Head of Low-Income Countries division in SPR of IMF
•    Brett House, Senior Fellow at CIGI (Centre of International Governance Innovation)
•    Yasemin Bal-Gunduz, IMF
•    Chair: Lucie Mboto, IMF

Presentations

Chris Lane, presentation of Occasional Paper released October 2013

Evolution of IMF programs and IMF engagement

LICS 2012: 20 small states, 8 net oil exporters, 27 fragile states, but the bulk of LICs by population are the ‘Core LICs’, of 30 countries with 765 million people

Four instruments used with LICs, 3 financial and one non-financial, from medium term support via Extended Credit Facility (most used, for Balance of Payments problems which can’t be resolved in a year or two). These can often be successive programs, and most are aid-dependent.

Short –term support comes from the standby facility, and for countries suffering shocks which can be resolved. A new instrument amongst this support is the rapid credit facility.

Non-financial support is for states to signal, via policy support instrument, including Uganda, Tanzania, Uganda.

These facilities, and their descendants, have been in place for 25 years.
The paper examines what is the impact of these facilities since circa 1986 when concessional financing began to be used (non-commercial).

Economic Impact of Fund programs
Literature on IMF programs is inconclusive as to the impact of Fund’s programmes in LICs on growth, though suggests it reduces inflation.

This study is important because it does not do what most of the literature has done, e.g. mixing episodic engagement in non LICs, so this looks only at longer-term engagement in LICs.

The key question to ask is how to control for the factors that distinguish countries which come to the Fund, which are defined by the fact that they face economic problems. The paper attempts to find a control group, e.g. countries with similar problems that did not have a program.

The broader context is that LICs’ economic performance has improved substantially in the last quarter century: all the boats have risen, so countries involved in the Fund or not are all likely to have improved.

This pattern is reflected also in reserves coverage which has massively improved, as has also been the case in terms of public debt, which can be accounted for via debt relief, better policies, and higher growth.

Our question was therefore not whether countries’ performance improved, but did it improve more in program countries. E.g. can we link program engagement to a strengthened improvement in performance?

Growth and poverty stylized facts reveal that average decadal real GDP per capita growth, and poverty gaps, shows that non LT engagement had worse improvement in growth and better improvement in poverty gap.

To estimate impact the model separated out longer-term engagement from short-term or episodic engagement, as longer-term engagement is often about signaling and institution building, rendering financing a component of the engagement.
Focus of presentation is on longer-term engagement.

Using fiscal and external position, aid-dependency, landlocked, political connections to major donors, presence of natural resource revenues, and the external conditions in the world, e.g. global upticks versus recessions.
This provided a group of countries which had and didn’t’ have programs, despite very similar conditions according to these prevailing factors.

A good match was Georgia which recent history was of intensive Fund engagement and Albania which lacks a history of Fund engagement.

The results reinforced the stylized facts of longer-term engagement, which was significantly higher GDP growth per capita, less volatility, lower inflation, much better improvement in fiscal balances, greater reductions in income inequality, and higher social spending, in particular significantly higher education spending

Short-term analysis, which was focused primarily on financing, with a less significant impact on growth but remained positive for large-shock impacts.

What explains better outcomes?
5 major channels
1)    policy advice and conditionality
2)    Debt relief
3)    Catalytic financing
These are only available via participating in a Fund programme
4)    Capacity Building
5)    Ensuring consistent macroeconomic framework
These channels are present in all countries, though the Fund believes these are more pronounced in program states.

Lessons for the Future

We expect more LICs to graduate in the future from being LICs or from the program.
This will lead to declining reliance on continuous IMF financing, with many countries graduating from using ECF as their balance of payments strengthens
Demand for non-financial instruments should increase, though shock financing will likely continue
Fragile state may still require repeated IMF support while they strengthen their macroeconomic positions

Given today’s announcement regarding the gold windfall commitment to the PRGT will allow roughly $2 billion annually at concessional rates, so that it won’t be hostage to member states’ approval.

Principal view is that the paper meets its remit, though that remit may have been very narrow.

The paper modestly characterizes its own innovativeness. This is important as to the distinction between longer-term and shorter-term lending. However the econometric basis of the pairing approach which finds that there is a success story may be more problematic.

It is not clear that the paper establishes that the benefit of the results is particularly due to the IMF’s engagement, e.g. due to the background of the commodity super cycle. Also the end of the cold war, leading to very different environments and relationships to ex colonial powers, regional powers.

Also in the wake e of the dotcom crash and to Lehmans there was a period of intensely cheap money, though the sovereigns haven’t accessed the cheap financing in these states their corporates will have, and that’s also not directly necessarily attributable to the IMF.

The major debt relief programs of the early 2000s, as much of the debt non-repayment means little impact was felt. The political economy of that period may be more significant.

To sum up, there is a great empirical review, which shows many good things have occurred in low income countries, but it’s not necessarily clear that there is a causal relationship. As we look forward it would b useful to think about how the results here dovetail with the recent strategic review at the World Bank which is targeting elimination of poverty and inclusive growth. These factors highlighted here are probably necessary for that, but are insufficient, and more factors probably need to be considered to account for achieve reduction in poverty.

Three interesting conclusions were found, that financing size may not have been so important, the modality of lending is important, and that there is a relation.

Here is where we need to examine the fund’s engagement historically. If financing is not so important, but more catalytic, then why do financing at all, and leave it to those who have an explicit developmental mandate, which the Fund lacks.

Regarding modalities, should there be such a thing as long-term engagement at all for the Fund? This is more the work of the bank and regional versions, though we’ve seen improvements, we should question whether the notion of long-term financing is successful is internal contradictory, otherwise it can’t be described as successful.

A follow-on paper would best serve answering which policies would be most critical? The Fund should be more bold in identifying what it thinks has been – and hasn’t – successful in supporting growth.

This currently suggests that the Washington Consensus fundamentally worked. AS such, how do we tailor those results/insights into specific cases. What are the things we need to do to support the different types of LICs?