The reform programme at its midpoint: sorting what worked from what was announced

Three years is the useful interval for assessing a reform programme. Early enough that the political conditions that produced it are still recognisable, late enough that formal adoption has had time to become implementation or to fail to.

Sorting the record into three categories makes the pattern visible.

Implemented and working

Electronic public procurement. Fully operational, mandatory, transparent, and generating measurable savings. It works because compliance is automatic — the system is the process, and there is no discretionary step where it can be circumvented quietly.

Banking sector resolution. Completed, painful, and the foundation of everything the financial system has done since.

Energy tariff reform. Implemented at very high political cost, and it removed the largest fiscal drain and the largest arbitrage opportunity in the economy simultaneously.

Central bank independence and inflation targeting. Established and maintained across changes of government.

Fiscal decentralisation. Underrated and among the most consequential. Local authorities received a defined share of tax revenue and spending responsibility, and the amalgamated communities that resulted became capable of doing things the previous structure could not.

Partially implemented

Public administration reform. New competitive recruitment procedures, higher salaries for some positions, and a genuine improvement in the quality of new entrants — alongside an unchanged institutional culture in most of the apparatus.

Health financing. The purchasing model was built and the primary care reform worked; hospital-level restructuring lagged.

Technical regulation harmonisation. Progressing steadily and slower than any timetable, with the constraint being conformity assessment infrastructure rather than legislation.

Anti-corruption architecture. The institutions exist, have statutory independence and produce investigations. Conviction rates for senior cases remained low, and the reason is the same one that appears in every other stalled reform.

Stalled

Judicial reform. Attempted through several distinct approaches — vetting, new courts, new appointment procedures — and blocked each time by the fact that the mechanisms for reforming the judiciary are administered by the judiciary.

Prosecutorial reform. Same structure, same outcome.

Privatisation of state enterprises. Repeatedly announced, minimally executed. The obstacle is that the enterprises generate rents for the people who would have to conduct the sale.

The pattern

Every reform in the first category could be implemented by building something new. Every reform in the third required the cooperation of people whose position it reduced.

That single distinction sorts the entire record, and it is a better predictor than the amount of external support, the strength of the government's mandate, or the technical quality of the legislation.

The implication for what comes next

A programme designed with this in mind would concentrate on measures that route around existing institutions and would treat institutional reform as a much longer-term project requiring generational change in the institutions themselves.

That is an uncomfortable conclusion, because the stalled reforms are precisely the ones that matter most for the rule of law and for investment security. But recognising which category a proposed reform falls into is the difference between a realistic expectation and a repeated disappointment.

For a company, the practical version is simpler: rely on the reforms in the first category, plan around the ones in the third, and do not price a Ukrainian investment on the assumption that court reform will arrive within the investment horizon.

Related in this archive

• ProZorro: what open procurement actually changes for a foreign supplier
• An election that was about the system itself, and what a mandate for change can and cannot do
• Elections and investment: what actually changes and what does not
• Ukraine Annual Review 2017: growth returns and the constraint changes

The simplest way to measure the gap between adoption and implementation is a business's daily experience: did the rule change, or only the text. The reforms that worked on my side always shared one feature — a record is kept and that record is open. Every reform without a record turns up three years later on the list of the stalled.