For decades, Nepalis have complained about a government that moves too slowly. Ministries work in silos, projects stall, databases do not talk to each other, and citizens struggle to get routine services. Now a young government seems determined to change that. Prime Minister Balendra Shah's administration has begun bringing digital governance, information systems, intelligence and monitoring of implementation closer to the Prime Minister's Office.
Should we welcome this, or worry about it? I think both questions deserve attention.
My first instinct was to welcome it. A prime minister needs reliable information to govern. If two ministries cannot share data, connecting them makes sense. If a road project has been stuck for years, a dashboard can show why.
But when information, technical talent and coordinating authority gather in one office, the center becomes more capable and also more powerful relative to everything around it. Information is not only capability. It is also power.
What others did with it
Tony Blair's government set up a Delivery Unit near Downing Street to track a few national priorities. The idea was to spread that discipline into the departments themselves. Singapore built common digital platforms near the center and has now set up an Institute of Digital Government that aims to equip more than 150,000 public officers with digital, data and AI skills by 2028.
That is the difference between a stronger center and a stronger state. If the center grows capable while ministries and local governments stay dependent on it, we have a stronger center. If it uses its temporary concentration of talent to make those institutions capable too, we have started building a stronger state. A good center should, over time, receive fewer problems, because its branches can handle them.
I have not seen a clear public explanation of which Nepal intends. Will technical teams be trained in ministries, provinces and palikas? How will anyone know whether capability is spreading beyond the center? After so many unstable governments, trust depends on those answers.
The other side of the coin
The same machinery that finds fraud can also let a state watch its citizens. China's connected databases show how information can move from administration toward monitoring behavior, though the systems are more fragmented than the popular picture suggests. In the United States, when Elon Musk's efficiency teams sought broad access to sensitive federal systems, courts, officials and states pushed back, and an April 2026 Government Accountability Office audit found that the Treasury had not fully followed its own security controls in giving a DOGE associate access.
Madison's old question
James Madison put the problem simply in Federalist No. 51: a government must first be enabled to control the governed, and then obliged to control itself. Elections are the primary control, he wrote, but experience shows the need for "auxiliary precautions."
What would those be in Nepal today? Who can see integrated government data, and is there a record of who looked? Can a citizen see what the government holds about them? Can even a prime minister be kept from unfettered access? Can independent constitutional bodies do their work without pressure?
That last question is not abstract. According to Nepali media reporting, senior officials of the anti-corruption commission spent hours at the Prime Minister's Office answering questions from advisers during the passport-contract investigation. Members of parliament later asked what had happened. But an independent constitutional body is not another delivery unit. Its independence matters most when its conclusions frustrate the executive.
A simple test
I am not questioning the intentions of the Prime Minister or his team. Try a thought experiment. Imagine the same digital systems, intelligence machinery and executive levers in the hands of a future prime minister you profoundly distrust. Would you still be comfortable with the arrangement?
It may seem an unfair question to put to today's government. But it is the question good institutional design requires, because systems outlast the people who build them. The real test is not how much authority the center can gather today. It is whether that concentration builds capable institutions throughout the country, with safeguards that work just as well under a prime minister we may not trust.
A dashboard in the war room can show a satellite image of an unfinished road in Rasuwa. It cannot hear the farmer still waiting for that road. A democratic state needs a way for that voice to reach the war room too.
A stronger center may help Nepal govern better today. A stronger state, one that works under whoever governs tomorrow, would be a much greater achievement.
Bohara is Emeritus Professor of Economics at the University of New Mexico. A longer version of this essay is forthcoming on his Nepal Unplugged Substack.