Selectivity on aid modality: Determinants of budget support from multilateral donors

Abstract

Since the late 1990s a selection on policy approach to aid was advocated such that more aid should be allocated to countries with good policies, but there is little evidence that this has occurred. This paper argues that donors may exercise selectivity over the aid modality. Specifically, multilateral donors will cede more recipient control over aid by granting more budget support to those recipients with better expenditure systems and spending preferences (towards the poor) aligned with the donor. We test this for European Commission and World Bank budget support over 1997–2009 and find some support. Both donors have given budget support to almost half of the countries they give aid, and it is usually a significant share of their aid. The principal determinants of receiving budget support are having a poverty reduction strategy in place, which can be considered a good indicator of aligned preferences, and indicators of government efficiency. These variables did not, however, influence the amount of budget support given. Multilateral donors have been more likely to give budget support to countries with aligned spending preferences and better quality systems, even if they have not reallocated the total aid envelope in that way.

Appendices

Appendices

An appendix with further detail on the data and supplementary estimates is available online at this journal’s webpage (Appendix B). The data used and Stata do files are available at https://sites.google.com/site/paulclist/data and on this journal’s webpage.

Appendix A: Proof, data and additional results

2.1 Appendix A1: Proof of proposition 1

The relevant constrains for the maximization problem in (A1) are: the incentive compatibility constraint for the $\underline \theta$ type (A4), the participation constraint for the $\bar{\theta }$ (A3) and the budget constraint (A6).

$$ \mathop {max}\limits_{{a_{BS}};{a_{PA}}} V :p\{ \lambda [\gamma {a_{BS}} - \underline \theta \underline q ] - \delta a_{BS}^2\} + (1 - p)\{ \lambda [\gamma {a_{PA}} - \bar{\theta }\bar{q}] - \delta a_{PA}^2\} $$

(A1)

$$ sub \quad \gamma {a_{BS}} - \underline \theta \underline q \geqslant 0 $$

(A2)

$$ \gamma {a_{PA}} - \bar{\theta }\bar{q} \geqslant 0 $$

(A3)

$$ \gamma {a_{BS}} - \underline \theta \underline q \geqslant \gamma {a_{PA}} - \underline \theta \bar{q} $$

(A4)

$$ \gamma {a_{PA}} - \bar{\theta }\bar{q} \geqslant \gamma {a_{BS}} - \bar{\theta }\underline q $$

(A5)

$$ A = p{a_{BS}} + (1 - p){a_{PA}} $$

(A6)

Given that (A6) is satisfied as an equality we can rewrite it as:

$$ {a_{BS}} = \frac{A}{p} - \frac{{(1 - p){a_{PA}}}}{p} $$

(A7)

Substituting (A7) into (A4) and rearranging we get:

$$ \frac{\gamma }{p}[A - {a_{PA}}] - \underline \theta \Delta q \geqslant 0 $$

(A8)

where $\Delta q = \underline q - \bar{q} > 0$. Substituting Eq. A7 into Eq. A1 we can then maximize with respect to ${a_{PA}}$, ${\mu_1}$ (the Lagrange multiplier associated with (A4)) and ${\mu_2}$ (the Lagrange multiplier associated with (A2)).

The first order conditions with respect to ${a_{PA}}$, ${\mu_1}$ and ${\mu_2}$ are given by:

$$ 2\delta \frac{{(1 - p)}}{p}[A - {a_{PA}}] - {\mu_1}\frac{\gamma }{p} + {\mu_2}\gamma = 0 $$

(A9)

$$ \frac{\gamma }{p}[A - {a_{PA}}] - \underline \theta \Delta q = 0 $$

(A10)

$$ \gamma {a_{PA}} - \bar{\theta }\bar{q} = 0 $$

(A11)

To analyze under which conditions ${\mu_1} > 0$ or ${\mu_1} = 0$ and ${\mu_2} > 0$ or ${\mu_2} = 0$ assume that ${\mu_1} = {\mu_2} = 0$. Then Eq. A9 can be rewritten as:

$$ A = {a_{PA}} $$

(A12)

Given $p \in (0,1)$, (A12) can never be realized and this implies that ${\mu_1} = {\mu_2} = 0$ can not be an equilibrium of the model.

Assume now ${\mu_2} = 0$. From Eq. A9 we then get:

$$ {\mu_1} = \frac{{2\delta (1 - p)}}{\gamma }[A - {a_{PA}}] $$

(A13)

Given $p \in (0,1)$ and the budget constraint being always satisfied as an equality, $A > {a_{PA}}$, which implies ${\mu_1} > 0$. The optimal marginal amount of ${a_{PA}}$ and ${a_{BS}}$ will be given by:

$$ a_{BS}^* = \frac{{(1 - p)\underline \theta \Delta q}}{\gamma } $$

(A14)

$$ a_{PA}^* = - \frac{{p\underline \theta \Delta q}}{\gamma } $$

(A15)

To check if the assumption of ${\mu_2} = 0$ still holds, rewrite (A9) as:

$$ {\mu_2} = {\mu_1}\frac{1}{p} - 2\delta \frac{{(1 - p)}}{{\gamma p}}[\tilde{A} - {a_{PA}}] $$

(A16)

Substituting the value for ${\mu_1}$ in (A13) into (A16) gives ${\mu_2} = 0$. Therefore, Eqs. 10 and 11 in the text are the final equilibria (it can be shown that the two remaining cases, ${\mu_1} = {\mu_2} > 0$ and ${\mu_1} = 0;{\mu_2} = > 0$, cannot be solutions for this model).

2.2 Appendix A2: Data sources

The data used for the analysis and the Stata files are available on this journal’s website and also at https://sites.google.com/site/paulclist/data. The data on GBS timing and value, total aid and number of donors are from the Creditor Reporting System (CRS) of the OECD-DAC. Information on the year in which a PRSP or IPRSP was agreed is taken from the IMF. All other data are from the World Bank. A detailed discussion of the data and sources is available in online Appendix B2.