The Australian Labor Party has said that, if elected, it will introduce legislation to specify the objectives of Australian aid, provide for rigour in its evaluation and ensure transparency in its reporting.

Labor’s mooted legislation could be expected to have something in common with the various bills for a Foreign Aid Accountability and Transparency Act that have recently been presented to, but so far failed to emerge from, the US Congress (the most recent is here). [Update: since this post was published, the US Foreign Aid Accountability and Transparency Act was passed and signed into law — read more here]. It might also borrow from the aid transparency and accountability legislation that has previously been enacted by several donor countries, or from relevant sections of other countries’ umbrella aid legislation.

This is not the first time that aid law of broadly this type has been proposed for Australia—the idea comes up from time to time, and appeals strongly to the Australian Greens. They pursued broad, UK-style legislation in 2013 and gender-related legislation in 2015, and their 2016 election platform includes a commitment to quite extensive legislation that would, among other things, establish an independent aid agency.

But is aid legislation needed at all? If so, for what in particular? If not, is it at least harmless?

A new Development Policy Centre Policy Brief  [pdf] examines precedents in Australian law and overseas, and argues that there is a case for legislation, provided it is tightly focused in certain ways and avoids inscribing aid allocation criteria and executive-branch administrative arrangements into law. Key areas for legislation should be transparency and monitoring and evaluation. However, no other donor to date has actually done a good job of translating sensible requirements in these areas into law.

Peer context: turning paper into rock

The main donor countries are divided, not quite equally, between those in which aid policy is controlled by the executive branch of government and those in which it is, to some significant degree, written into law and therefore at least theoretically constrained by the legislature. The most significant pieces of overseas legislation are collected here, and summarised in the Policy Brief.

Among the four largest aid donors, the United States and the United Kingdom have plenty of aid law; Japan and Germany have none of any consequence. Of the 21 main OECD donors, nine have enacted some significant piece of aid legislation.[1] Most recently, the governments of France and Italy both put in place overarching legislative frameworks for their aid efforts in 2014, and the UK government legislated its commitment to meet the UN’s 0.7% target for Official Development Assistance (ODA) as a proportion of Gross National Income (GNI) from 2015.

No matter that it might be written in stone, it is rare for overarching aid legislation to have any more bite than white papers or other policy and strategy documents. Typically, such legislation sets out broad objectives and operating principles, gives some indication of geographic, sectoral and thematic priorities and says something about processes, particularly those relating to accountability and reporting. Where it is not general or indeed vague, it is typically equipped with escape hatches. The UK’s 2015 Official Development Assistance (ODA) target legislation does have substantial bite, as it creates a ‘duty’ on the part of the Secretary of State for International Development to meet the 0.7% ODA/GNI target. The UK’s 2002 umbrella legislation has some bite too, but rather less. While it is often said that this legislation makes it ‘illegal’ to use British aid for purposes other than poverty reduction, the law says only that the Secretary of State must be ‘satisfied that the provision of the assistance is likely to contribute to a reduction in poverty’; Canada’s 2008 aid legislation contains similar language.[2]

Australian aid law

If 12 of the 21 major donors get by well enough without aid law, and it is not doing a great deal of work where it does exist, what is it—or might it be—good for in the Australian context?

In Australia, and in fact in most countries at most times, legislation can usually be described as having one of two main functions. Either it relinquishes a morsel of executive power, for example by passing it to an independent agency of some kind or submitting to the obligations associated with international agreements, or else it defines certain actions or omissions as illegal such that they attract specified sanctions.[3]

Australia does not use legislation to create or dismember ordinary government departments and agencies—just as well given that machinery-of-government changes are frequent and extensive. For example, no change in legislation was required to abolish the Australian Agency for International Development (AusAID) in 2013; that was done by executive order. However, Australian governments have enacted aid-related legislation from time to time for specific purposes (leaving aside routine budget legislation). Most notably, legislation was passed in 1982 to establish the Australian Centre for International Agricultural Research (ACIAR), a statutory authority within the Foreign Affairs and Trade portfolio. And legislation has been passed as needed to formalise Australia’s membership of international organisations, such as the World Bank’s International Development Association (1960) and, most recently, the Asian Infrastructure Investment Bank (2015).[4]

For what then, exactly, is the Labor Party proposing to legislate? Labor appears to be offering about the same menu of items that one finds in aid legislation in most of the nine other countries that have it: objectives, themes, processes relating to transparency and accountability, and evaluation arrangements. That the emphasis in Tanya Plibersek’s October 2015 speech was on transparency and accountability, rather than objectives and themes, might or might not be significant. If something akin to the UK’s or Canada’s vanilla legislation on transparency, accountability and reporting is intended, then it is not very significant. If something closer to the proposed US legislation is intended, then it is significant. Whatever the details, any umbrella legislation along the lines outlines by Plibersek would be a first for Australia’s aid program.

Perils and pitfalls

Given that Labor’s proposed legislation runs on familiar rails, it faces familiar pitfalls. There are two main problems with manufacturing overarching aid legislation. One is that once you start, you can’t be sure where you will stop. A second and perhaps more insidious problem is that, where it makes stronger demands on aid agencies than on other agencies of government, aid legislation tends to embody a particular distrust not only of the executive, present or future, but of the very notion of aid.

Given these problems, it is tempting to conclude that aid law should be confined to specific purposes—formalising international agreements, and occasionally creating development-related institutions with statutory independence (it is conceivable, for example, that future governments might wish to create an Australian Overseas Private Investment Corporation, or an Australian Independent Commission for Aid Impact.)

The contrary view is that aid is special—both unusually important if considered in terms of potential impact per dollar spent, and unusually vulnerable to the vagaries of political change and fiscal fortune. In a democracy, on this argument, it is is perfectly legitimate to deploy legislation as a tool with which right-thinking present governments can tie the hands, in the nicest possible way, of wrong-thinking future ones. This concept of lashing governments to the mast is presumably what underlay the act of legislating 0.7% in the UK.

Whether and how far aid law does tie the hands of governments depends on the structure and composition of the parliaments concerned. And when the ties do come off, sooner or later, there is a clear risk that all that chafing gives rise to a disproportionate force in the opposite direction. Already we might have seen something like this in Australia in connection with the conferring of Executive Agency status on AusAID in 2010, which was not even a legislative act. The integration, or indeed disintegration, of AusAID into DFAT three years later can be read as an over-correction in response to perceived over-reach.

The converse risk, admittedly a lesser one in terms of consequences, is that fear of backlash or the difficulty of negotiating a bill through parliament leads to the enactment of a rosy, vague and pointless piece of legislation—something that nobody would wish to spend political capital in repealing or amending, but also something that has no practical effect.

What’s aid law good for?—Fundamental, apolitical things

If Labor or for that matter any party wants to legislate usefully about aid, and if it wants the legislation to be both durable and of enduring relevance, international experience suggests several ‘don’ts’. First, don’t legislate objectives, themes or geographic and sectoral priorities; leave those in the domain of white papers and other policy documents. Second, don’t legislate the details of mundane monitoring, reporting and evaluation processes; these things reliably happen anyway thanks to international norms and political defensiveness and some reporting already has a sufficient basis in other legislation. And third, don’t legislate administrative arrangements except where statutory independence is an inherent requirement of an agency’s mandate. In particular, don’t legislate for independent evaluation unless the plan is, for better or worse, to create a statutorily independent aid evaluation agency.[5] ‘Independent’ is just too vague a term.

But what about the ‘dos’? With an eye to the proposed US bill’s merits but also its defects, our view is that there is a case for the introduction—by any government—of Australian aid legislation incorporating the following three requirements.

1. All activities above some threshold size should be reviewed before the mid-point of their lives; larger and experimental activities should also be evaluated on completion;
2. specific, current information about aid-funded activities and partnerships administered by any government agency, including activity-level financial information, reviews, evaluations, designs and other project documents, should be made publicly available in accordance with the IATI Standard within a short, fixed period of time after its production; and
3. the Australian National Audit Office (ANAO) should be required regularly to assess the extent to which relevant government agencies have complied with the requirements at (a) and (b).

Both monitoring and evaluation and transparency, unlike aid objectives, themes and allocation criteria, are essentially apolitical topics. Legislating clear requirements in these areas (and including a role for the ANAO in assessing compliance) would be highly unlikely to lead to any backlash, and no government is likely to want to repeal or water down such legislation once it has been passed.

More importantly, legislating requirements in these two linked areas could be expected to have beneficial flow-on effects in other areas. Once it is possible for anybody to gain a clear and current view of what is happening down to activity level in the aid program, and also to see how effective activities and programs are assessed to be in expert reviews and evaluations, improvements can be expected to follow in areas such as priority-setting, resource allocation, the management of poor activity performance and the alignment of programs with partner government priorities. If reviews or evaluations are tending to be whitewashes, this will quickly be perceived. Legislating for searching scrutiny is not the same thing as legislating for effective aid, but it might well be the best proxy.

Jo Spratt responds to this post here.

Robin Davies is the Associate Director of, and Camilla Burkot a Research Officer at, the Development Policy Centre.

---

[1] Among the 21 donors in question, the nine with significant aid legislation are Austria, Belgium, Canada, France, Italy, Korea, Luxembourg, the United Kingdom and the United States. Seven very small donors, who have only recently joined the OECD Development Assistance Committee, are not considered here.

[2] Canada’s legislation says, ‘Official development assistance may be provided only if the competent minister is of the opinion that it (a) contributes to poverty reduction; (b) takes into account the perspectives of the poor; and (c) is consistent with international human rights standards.’

[3] Of the nine donor countries with significant aid legislation, two have used it mainly to create new administrative structures. Austria legislated in 2002 to create the Austrian Development Agency as a non-profit company. Italy’s 2014 legislation is quite broad in coverage but its main effect is to create the autonomous Italian Agency for Development Cooperation.

[4] One might expect such legislation to be repealed when Australia cancels membership of an international organisation, as happened in the case of the International Fund for Agricultural Development. Australia announced its decision to withdraw from the organisation in 2004 and the decision took effect in 2007. However, the 1977 legislation formalising Australia’s membership remains in force.

[5] There are arguments for this but also arguments against, based in both logic and experience. Too much distance between an aid agency and its evaluator, it is often argued, degrades the quality of evaluation and the value of evaluation results.