Is debt relief aid?

UK charity Action Aid has been getting quite a bit of press

today with its report

that only one-third of G7 overseas development assistance (ODA) is "real"

aid. The rest, they say, is "phantom aid". (If you’re interested,

the full report is here,

in PDF format.)

The thing that struck me about the report was where Action Aid said that debt

relief counts as "phantom aid". I think that debt relief is hugely

important for poverty reduction, and so I was rather angry abou that, especially

since Action Aid defines "phantom aid" as "aid which may have

achieved other goals, but did not help to fight poverty".

There are many, many things to take issue with in the Action Aid report –

for instance, since only 30% of the world’s poor live in middle-income countries,

Action Aid counts any aid to those countries beyond 30% of the total budget

as "phantom aid". This is profoundly silly: poverty-reduction programs

are, if anything, more effective in middle-income countries, because

those countries have better institutions.

But let’s stick with debt relief. Consider a highly-indebted country with a

poverty problem. It needs to spend money on providing clean water to its poor,

but it also needs to spend money on debt service. In this situation, by far

the most efficient way for a foreign government to get clean water to the people

who need it is to simply forgive that country’s debts. All the transaction costs

of the foreign government trying to provide the water itself – from consultants

to bureaucratic procurement procedures – disappear at a stroke. The local

government, which is certainly better placed to provide water than the foreign

government is, gets budget money freed up to be spent where it’s needed.

The local government can also, with its newfound creditworthiness, start to

finance projects which are needed for the long-term health and growth of the

nation, like building roads or ports. What’s more, it can quite easily reallocate

money where it’s needed when it’s needed, something a foreign government would

find almost impossible.

Now, debt relief on its own does not necessarily decrease poverty. If the money

saved in debt service is spent instead on private jets for the president and

kickbacks to his buddies, then the debt forgiveness will have done no good at

all. That’s one reason why the HIPC

program was so slow to get off the ground: it took a long time for governments

to prove that they were worthy of it. But when it’s done right, debt relief

is pretty much the most efficient way of delivering aid imaginable. Just think

of the costs involved with the average World Bank poverty reduction program,

complete with environmental impact assessments and whatnot, compared to the

cost of a local government simply going out and doing what is needed for its

poorest citizens. So I’m pretty upset that Action Aid simply declares that no

debt relief helps to fight poverty.

Then, however, I read the Action Aid report, and it does make some good points.

Here’s what it says:

All debt relief provided since 2002 has been counted as part of ODA,

despite the fact that the Monterrey Consensus agreed that year explicitly

stated that aid increases should be additional to debt relief.

Irrelevant. While the Monterrey Consensus never said that

debt relief should be excluded from ODA budgets, it did set a target for ODA

which explicitly excluded debt relief. To no one’s great surprise, many countries

are a very long way from reaching that target, even when you include

debt relief. But debt relief is a type of aid, and it is therefore perfectly

reasonable to consider it ODA. If Action Aid wants to castigate countries for

not meeting their Monterrey Consensus goals, that’s fine. But it does not mean

that debt relief is "phantom aid".

Cancelled debt stock – the principal and interest on the loan –

are counted as ODA in the year in which the relief is agreed, even though

any benefits are felt over several years.

Arguable. On its face, this is quite a strong argument. Let’s

say that Freedonia is paying $5 million a year on $100 million of debt, and

that Belgium then decides to forgive that debt. Suddenly, Belgium’s ODA expenditure

for that year has gone up by more than $100 million, while the amount of cashflow

freed up for poverty reduction is a mere $5 million. Clearly, Freedonia would

have been better off simply taking the $100 million from Belgium, using $5 million

of it to service the debt, and spending the rest on poverty reduction.

On the other hand, there are two reasons why debt relief is accounted for this

way. The main one is that this is simply the way that budgets in the OECD work.

Loans are assets, and accounted for as such. When a loan is written off, that’s

a line item in the country’s budget, and the full amount of money needs to be

found for it immediately.

Furthermore, Freedonia gets more than just a cashflow benefit from the write-off:

it also gets a significant decrease in its indebtedness. Without a write-off,

Freedonia will simply continue to pay interest on its loans and never pay down

the principal. With the write-off, Freedonia’s debt burden falls substantially,

making the country’s economy more stable and attractive to investors.

These figures exaggerate the actual transfer being made to poor countries

because debt relief is valued at its full nominal value. Much of the debt

relief provided to poor countries simply closes the gap between what countries

were scheduled to repay and what they actually were able to repay, and has

often done little to relieve budgetary pressure on poor countries.

Fair. If Belgium had already agreed that Freedonia need only

pay $1 million a year on its debt, then claiming over $100 million in ODA upon

forgiveness seems even more exaggerated. Freedonia gets ony $1 million a year

in immediate cashflow benefits, while Belgium gets to crow about its enormous

ODA budget. There’s no way that Belgium’s action can be spun as meaning $100

million’s worth of poverty reduction.

What’s more, Action Aid shows that total debt service costs, especially in HIPC

countries, have actually been rising, rather than falling, even as large chunks

of their debt have been forgiven. It’s good that the debt has been forgiven,

of course. But without cashflow benefits, it’s not clear that there’s been much

poverty reduction as a result. And it certainly seems unlikely that even a small

fraction of the $9.4 billion cancelled in 2003 made its way to direct poverty-reduction

programs.

In the UK, debt cancellation has been presented as additional to aid

spending. This is double counting.

Unfair.The UK, as far as I know, has been quite upfront about

its debt relief activities and their place within the ODA budget. No one’s counted

debt relief on its own and then added it to an ODA total which already includes

it. Therefore, no double counting. The Action Aid methodology says that all

debt relief is "phantom aid", not just debt relief which is double-counted.

Funding debt relief from aid budgets is not only misleading. It also

risks penalising countries that are not indebted, as aid resources are diverted

towards heavily indebted countries. It also violates the principle that creditors

should carry some of the cost of debt relief, given the role that reckless

lending has played in the debt crisis, and the fact that much of the initial

lending was not supporting development-related expenditures.

Arguable. While it might be fair to include debt

relief within an ODA budget, it’s not fair to fund debt relief

with an ODA budget – and I daresay more than one country has

been doing that of late. In other words, certain countries might have spent

money directly on poverty reduction which they allocated instead to debt relief.

That’s a bad thing. The big question is the extent to which that behaviour goes

on, and since it’s all based on a hypothetical – "how big would your

ODA budget have been had you not done the debt relief?" – it’s very

difficult to answer.

Poor countries without debt have not necessarily been penalised – that’s

a strong word – but it’s true they haven’t had the same amount of attention

paid to them that the HIPC countries have had.

As for the "principle that creditors should carry some of the cost of debt

relief", aren’t the creditors carrying all of the cost of debt

relief? If they’re not, who is? I think I understand what Action Aid is trying

to say here, but it’s a very weak argument indeed.

Weirdly, after running through its arguments, Action Aid concedes my main point:

Debt relief can be a particularly effective form of resource transfer,

as it is untied, stable, predictable and flexible.

Well, exactly! The only argument seems to be whether it belongs in ODA budgets

– and if it does turn up there, whether it is "phantom aid".

My view is that any effective, stable, predictable and flexible form of resource

transfer is most definitely foreign aid, and not phantom aid.

I think the biggest problem with the Action Aid report is that there’s an unspoken

assumption running throughout it that we generally think that the amount of

poverty reduction is proportional to the amount of money that is spent on poverty

reduction. Action Aid gets halfway to the truth of the matter: that some parts

of the broad global poverty reduction program are much less cost-effective than

others. But the charity also confuses the matter by trying to draw a bright

clear line between "real aid" and "phantom aid".

In fact, all aid is useful to some degree, and there is no such bright clear

line. Countries spend money on ODA for many reasons, and some of that money

is always going to be more cost-effective, in terms of poverty reduction, than

other chunks of that money. The problem is that Action Aid spends a large amount

of space in the report comparing the amount actually spent on aid to the amount

that it thinks should be spent on aid. It’s confusing means (money) with ends

(poverty reduction).

In fact, there’s no formula tying a certain number of billions of dollars to

a certain number of lives brought out of poverty. Action Aid shouldn’t be concentrating

on dollars, it should be concentrating on people. Scaring up a "scandal"

(cf the Guardian headline) over "phantom aid" doesn’t achieve that

in the slightest – if anything it reduces pressure on governments to spend

more and do more. After all, if two-thirds of the money is wasted, what’s the

point of spending the money at all?