Digital connectivity is viewed by many in technology and development circles as central to achieving the 17 Sustainable Development Goals (SDG’s). Luminaries supporting the Connectivity Declaration are among them.

Technology can indeed be transformative. India’s digital identification scheme (Aadhar) has given over 950 million Indians a verifiable identity, enabling millions of poor people to receive government subsidies that were previously being diverted to others or for which they had to pay bribes, also saving the government billions of dollars.[1] But such examples are rare. Governments invest billions of dollars annually in automating government financial management, customs, and tax systems respectively. But aside from a few cases that are repeatedly quoted, there is little rigorous evidence of the impact of these investments on government efficiency and accountability to deliver services.

Ironic, given the abundance of data in the digital age.

The World Development Report 2016 “Digital Dividends” attempts to fill this gap and assess the impact of e-government. The evidence is decidedly mixed. E-government systems in general do increase budget transparency (figure below), but e-filing of taxes and e-procurement, among the most common applications of digital technology, have ambiguous effects on tax compliance, transparency, and perceptions of corruption. E-filing of taxes is meant to reduce the administrative burden to businesses and citizens by minimizing interactions with tax officials and reducing opportunities for rent-seeking. But data on firms’ perceptions in Europe and Central Asia (ECA) from the World Bank Enterprise Survey finds that the introduction of e-filing, on average, does not reduce the number of visits of tax officials to firms. E-procurement is meant to increase transparency and competitiveness of government procurement. But firms in ECA on average are no more likely to bid for contracts, nor report any reduction in the likelihood of being solicited for bribes to secure these contracts, after the introduction of e-procurement.

These mixed results are symptomatic of a broader, sobering trend: many government digital technology projects fail. An examination of the roughly 530 e-government projects funded by the World Bank in the past twenty years (Figure below) for example, reveal that almost a third are unsatisfactory, and perform worse than a typical project. ICT project failure is a recurring theme in audit reports of the governments of the US, UK, and many other high income countries.

The main takeaway from the WDR 2016 is that digital technologies can indeed be transformative; but only if accompanied by “analog complements”. Public officials must have the incentive to effectively use the technology, which in turn depends on the strength of government institutions. In patronage-based bureaucracies, for example, officials will resist e-government as the automation of business processes and improved monitoring reduce bureaucratic discretion and opportunities for petty corruption. Firm perceptions of tax compliance improved in the ECA countries where e-filing was accompanied with simplified tax regulations and improved management practices in the tax authority. World Bank-funded e-government projects perform better in countries with better quality institutions (as measured by the Worldwide Governance Indicators, and controlling for income).

To quote Bill Gates “automation applied to an efficient operation will magnify the efficiency … automation applied to an inefficient operation will magnify the inefficiency.”

[1] Barnwal, P. (2015). Curbing leakage in public programs with biometric identification systems: Evidence from India’s fuel subsidies. Working Paper. Muralidharan, K., Niehaus, P. and Sukhtankar, S. (2014). Building state capacity: Evidence from biometric smartcards in India. NBER Working Paper No. 19999.