This weekend, the United Nations officially adopted the Sustainable Development Goals (SDGs), or Global Goals, as they now seem to be called. Judging from the congratulations emanating from New York, this is great news for the international aid community, who managed to replace the Millennium Development Goals (MDGs) with a new framework to cover the next 15 years until 2030. Unfortunately, they’re also bad news for developing countries.

It’s been widely reported that the process of tortured negotiations essentially turned into what political scientists call “logrolling”: everyone voted for everyone else’s proposal to garner support for their own pet causes. The result is the current unwieldy framework of 17 goals, 169 targets and probably an immeasurable number of sub-targets and indicators. There are many ways in which this result could be, and has been, criticised. At last year’s ODI Centre for Aid and Public Expenditure Conference, CGD’s Charles Kenny called the SDGs “a mess”. Oxfam’s Duncan Green bemoans the technocratic hubris of the process, which involved every conceivable NGO lobbying for their cause without concern for the whole edifice, among other things. Ex-World Banker Phyllis Pomerantz makes the rather obvious point that if the world has 169 priorities, it effectively has none. Of course all of these arguments stand against an avalanche of enthusiasm, perhaps best encapsulated by Bono quoting Nelson Mandela: “it’s always impossible until it’s done”.

The 17 goals cover very broad swathes of government activity and explicitly aim to “transform societies” in the pursuit of ending hunger and poverty, improving health care, education and water and sanitation services, creating welfare states and peace in our time, and much more. The drafters of the SGDs understood that the global volume of aid would fall far short of funding much of this agenda, which is why the outcome document of the recent “Financing for Development” conference in effect calls for developing countries’ (hypothetical) domestic revenues to be used to pay for it all.

Unfortunately, the drafters of the SDGs cannot possibly claim a mandate for an unprecedented worldwide transformation, nor is there any evidence that implementing them by means of a 169-goal strong global plan has any chance of succeeding.

Let’s take a step back. Technically, the SDGs and its implementation framework are an agreement among the sovereign states that make up the membership of the UN. Agreements between states are of course as old as states themselves, but their success, among many other things, depends on how deeply they rely on the multitude of domestic institutions and power relationships to produce a specific result.

If the agreement only concerns how states behave internationally, implementation is relatively straightforward – think a peace agreement, where the governments involved agree to stop fighting each other. Regulatory agreements, like the Montreal Protocol, are much harder to implement, since they rely on many states taking action domestically in very specific ways. Institutional agreements, where the states at the table decide to change their very nature, are exceedingly rare, and have virtually no record of working without the threat or use of force. There is, for instance, the Holy Alliance of 1815-1825. Scared by the French Revolution, its members agreed to remain monarchies and fight to keep Europe that way. The Comecon Treaty is another example, by which the member states of the Soviet Bloc agreed to organise themselves as socialist economies and form an integrated trade system. If these examples sound obscure and a bit scary, that is the point. Yet they pale in comparison to the scope, scale and specificity with which the SDGs aim to reorganise the inner workings of every country in the world.

It is impossible for the complex societies that make up the countries of the world to be so perfectly aligned that they would be able to agree on so much at the same time. In various countries, groups will inevitably have preferences that differ from the SDG agenda. And we know from the comparatively simple and focused MDGs that misallocations, or “kinky development”, can be a real problem. Even a policy as seductively simple as universal access to primary health care can be deeply flawed, as Jeffrey Hammer persuasively pointed out.

Furthermore, the likely futile efforts of aid donors to set up new and expanded health, education, social protection and other services will harm the ability of developing countries to establish such systems on their own. There is ample evidence, for instance, of how Poverty Reduction Strategies (the main vehicle to try and link MDG funding to government systems) resulted in costly parallel systems, wasted efforts and very little positive effect on actual government capabilities. The current momentum for a “data revolution” to support the SDGs promises more of the same.

In reality, governments will need to continue the slow and difficult process of institutional transformation if they want to develop – they need to build capable bureaucracies, money management systems, train and deploy public servants and so on. The way to do that is for government officials and politicians to fix the problems that are in front of them, because they want to keep their jobs and the complicated networks of domestic politics push them that way. Much like Bismarck, who set up the German welfare state to keep the socialists out of government, this kind of progress need not rely on politicians wanting to the right things, and is intensely contested by its very nature.

Governments develop unevenly, and find their own idiosyncratic ways to do so. A good way of organising a welfare system in one country may look very different from another, and societies at different times will prioritise some areas over others. Arguably, this is how most progress has been achieved on the MDGs as well – not by the UN’s plan, but because governments were advancing anyway. China probably doesn’t mind the international plaudits, but it is hard to imagine that its policymakers paid much heed to the international agenda when they were busy achieving all that growth in the 1990s and 2000s, thus lifting unprecedented numbers out of poverty.

For those governments doing the actual work of development, capacity is scarce, and every step forward is often a struggle, especially in poor and fragile contexts. They can ill spare the time and attention to deal with a new set of measurement frameworks, assessments, strategies and plans – ones that now aim to not just deliver aid but impound their own tax revenues in the process. To them, the SDGs really are a Chimera, in both senses of the word. An imaginary monster compounded of incongruous parts, as well as an unrealisable dream.

Philipp Krause specialises in public administration and budgeting, particularly ministries of finance. He leads the public finance team at ODI’s Centre for Aid and Public Expenditure and manages the research program of the Budget Strengthening Initiative.