You probably don’t need to be a memory champion to remember the last time you heard, saw, thought, or said the word “accountability.” If you work in or on international development, it was probably this week—maybe even today.
Lant Pritchett and I argue in our new paper that accountability is, indeed, central to high-performing governance and welfare-enhancing service provision (education, health, etc.) systems.
Strengthening accountability is often taken to mean strengthening the reporting of what can be observed, verified, quantified—what we call “accounting-based accountability.” Accounting-based accountability solutions will have substantial impacts in improving the efficiency and efficacy of governments in carrying out activities that are, by their analytical nature, logistical. There may well be some public-sector bureaus where accounting-based accountability is likely to improve performance (e.g., post office delivery or vaccine delivery).
But most of what government does isn’t like that; and, indeed, for high-functioning organizations which do things that aren’t fully logistical, “accountability” means much more than accounting-based accountability. Accountability in these organizations is “account-based,” focused on a justifying analysis or explanation. Many accounts will involve the hard numbers of accounting, but the account and the accounting are not the same thing. The accounting is an input into a broader account of performance.
It matters that we misuse “accountability” in practice. Yes, this is by definition a semantic point. But it is a semantic point with very tangible real-world implications. A lot of interventions to strengthen accountability rely on strengthening accounting. And if we’re right then even though accountability is central to good performance, accounting-based accountability interventions aren’t likely to be good strategies for improving system performance.
Tech won’t (often) solve accountability problems
Technology can do incredible things—but solving accountability problems isn’t likely to be one of them. ICT solutions, from cameras to data management systems, primarily strengthen accounting-based accountability, which in our view is unlikely to work. We explore this in the Education sector, an area where Accountability ICT is often trumpeted.
The key elements of what teachers do is not amenable to the accounting-based accountability which Accountability ICT can strengthen. An education system that is centered around learning needs to deploy teachers who have adequate resources and capacities, and who act with sincere concern for the learning progress of each child. It is hard to reduce teaching to a list of “hard” characteristics (e.g., has a master’s degree) or “thin” behaviors (e.g., is in attendance). “Acting with concern,” much less “acting with sincere concern,” for each child is not amenable to improvement with accounting-based accountability.
There are good reasons to worry that a focus on accounting-based accountability may not just be neutral, but actually detrimental, to system-wide performance. If people working in the education system invest time in collecting what can be measured and reported, they will do so by investing less time in what cannot be as easily accounted for—like teaching quality, individual instruction, relationship-building, responsiveness to parents or other teachers, etc. If reporting becomes a tool of control it can constrain teacher autonomy, leading some of the best teachers to exit. Top-down control using accounting-based accountability channels may also demotivate teachers from learning about their students if the teachers cannot make productive use of that information.
Education—and much of what matters in the public sector—is not amenable to Moore’s Law, the exponential increase in the power, and decrease in the relative cost, of computing. There is no evidence whatsoever (we attempt to show) that improvements in technology (and thus the availability or cost of Accountability ICT) are, or have ever been, the binding constraint on high-performing education systems. All OECD countries managed to create high learning performance education systems before ICT for accountability was an option.
So what do we do?
What is the alternative to Accountability ICT, and a focus on strengthening accounting-based accountability? Our answer is to focus on building locally embedded account-based accountability (both horizontally to peers and downwards to communities) in education and beyond.
Some, on reading this, may roll their eyes, imagining interventions unlikely to achieve scale. The scalability of accounting-based accountability is seductive; but it is a siren song that will often lead to shipwrecks on the long voyage of development. To rapidly implement at scale accounting-based accountability where it is ineffective, or even deleterious, is to accomplish nothing. In addition, to say that each intervention must be “deep”—digging into the governance particulars in every given setting—does not mean that it cannot be broad, too. That is, there is no reason that a push for more account-based accountability cannot be widely pursued in education reform initiatives throughout the developing world.
Neither accounting nor technology ought to disappear. Some of the metrics which are the focus of so much energy ought to stay. These metrics shouldn’t be seen as answers to performance and accountability questions, but rather as inputs into a broader process of making an account.
Much of what is currently fashionable in education reform assumes that external catalysts—be they reformers from the capital or outside agencies—are going to provide critical transformative inputs. We think this is unlikely, as it misunderstands both the nature of the problem (a governance and accountability challenge, first and foremost) and the most likely class of solution (strengthening account-based accountability).
The pyrrhic victories of accounting-based accountability
Even where accounting-based accountability solutions show results they do so by advancing progress on a dead-end road, one which terminates far below the desired summit of transformed systems. In the paper we argue this point using the documented successes of cameras in classrooms.
Even if cameras are the most straightforward way to improve teacher attendance, they are in our view the wrong one. This is because a teacher who shows up because they are being observed may add value over an absent teacher, but is not likely to do all the other (unmonitorable) things needed to produce high levels of student learning and growth. And teachers that are likely to produce high levels of student learning and growth are likely to be repelled by a system that focuses on accounting-based accountability and top-down control and observation.
The road to a high-functioning education system not only addresses teacher attendance but also what happens when teaches do show up. Education systems steeped in accounting-based accountability will sometimes be better than the status quo, but won’t be able to deliver the education all children deserve. A misplaced focus on accounting-based accountability moves us away from, not towards, our broader goals. Systems transformation in education and far beyond depends on our ability to differentiate account-ability from accounting ability.