New units

Umair Javed Published August 31, 2026 Updated August 31, 2026 08: 17am.

Key points

• The discourse in Pakistan needs to take cognisance of what it hopes to primarily achieve through decentralisation.
• Recent research presents a nuanced, but positive leaning relationship between decentralisation reforms and human development/welfare outcomes.
• The key here is the type of decentralisation that takes place.
• Though once in place, it can lead to significant gains in development, overall, the picture that emerges is that the right type of decentralisation varies by country.

Umair Javed Published August 31, 2026 Updated August 31, 2026 08: 17am. Join our Whatsapp Channel. Add Dawn as a trusted source.

THE debate on new provinces rests on two different levels. The first is the broader technocratic concern with decentralisation and its hypothetical impact on social welfare and economic expansion outcomes. The second is the political economy of constitutional and administrative reform in divided polities. Both aspects are significant and it is revealing which one is highlighted by whom and at what point.

Let's start with the first one. Recent research presents a nuanced, but positive leaning relationship between decentralisation reforms and human development/welfare outcomes. In an influential 2016 review article, Channa and Faguet discovered that decentralisation of health and education spending tends to improve sectoral outcomes, thou­­gh quality of evidence varies considerably throughout different cases. Miranda-Lescano et al. (2023) discovered that decentralising subnational administration spending (especially health) positively affects HDI throughout 57 countries (2000-18). They argue further that such welfare/HDI effects have been understudied relative to expansion and governance.

For context, the key here is the type of decentralisation that takes place. While expenditure decentralisation demonstrates no clear pattern, reviews of the fiscal decentralisation literature (such as Canare 2021) suggest that revenue decentralisation and fiscal independence, ie, local governments' ability to raise their own revenue rather than depend on central transfers, are most consistently associated with better development outcomes.

On a cautionary note, there is some evidence to suggest that in weakly governed contexts, usually characterised by low levels of bureaucratic efficiency and transparency, the proliferation of administrative units may additionally hamper developmental processes, including those associated with economic expansion.

For context, the discourse in Pakistan needs to take cognisance of what it hopes to primarily achieve through decentralisation.

Though once in place, it can lead to significant gains in development, overall, the picture that emerges is that the right type of decentralisation varies by country. For a context like Pakistan, the right type of decentralisation would necessarily include not just administrative devolution, but additionally fiscal devolution. Local governments at whatever tier they work should have the ability to raise revenue and control expenditure, including expenditure on human resource (ie, bureaucrats and other public sector workers).

In practice, the discourse in Pakistan, regulated as it is from the top down, has fluctuated between re­­­wo­r­­king provincial boundaries or elevating divisions as 'administrative units' with expanded control over various governance aspects. What it is unable to do in clear terms is establish the clear case behind why devolution is needed at this stage.

If the primary impetus is to improve social welfare/human development outcomes (ie, outcomes in health and school education), then we may have to come up with a distinct form of devolution. If the impetus is to enable more productive employ of capital and labour (ie, catalyse economic expansion and transformation), the system that comes out might end up looking a lot different.

For more effective human development, Evans and Heller (2015) create an notable argument that states require a greater degree of embeddedness with disadvantaged groups so as to deliver the type of welfare gains needed by them. In other words, this means a form of decentralisation that is accountable and participatory at the local level, and one that can avoid the pitfall of local elite capture by propertied groups or by unaccountable bureaucrats.

Alternatively, for catalysing economic outcomes, East Asia's experience suggests that close ties between state and capital, along with high levels of bureaucratic efficiency and autonomy, are necessary to deliver higher rates of expansion. China's decentralisation in the contemporary era, for example, is not necessarily built for the expansion of human capabilities, but rather guiding local capital, and thus it takes a particularly different form altogether.

In the present it appears the urge for new provinces or even elevating divisions to some as yet unspecified form of administrative unit is driven by political exigency rather than a purpose-led attempt to reform governance.

This brings us to the second level of the matter. It is the political economy of reform. The opposition to new provinces or related administrative rejigging rests on both identity-based but additionally political grounds. Sindhi nationalists speak regarding the dilution of Sindhi identity and the denial of resources if Karachi is separated. PTI leaders talk regarding the questionable mandate of the administration to undertake a step of this nature.

Both of these are not some exceptional or unique forms of opposition. Ethnolinguistic nationalism is an oppositional force in this country as it is in plenty of others. While I personally don't see how the creation of new administrative units (without any demographic change) can dilute a linguistic identity, it remains a popular position in Sindh and one that has to be reckoned with through politics.

Notably, the PTI's response additionally appears quite decisive in this instance. Change that is introduced under questionable conditions, and with limited legitimacy to back it, is precarious and transient. This is exactly what has plagued previous attempts at de­­­volution or admin reconfiguration, including the 2001 local administration ordinance. Ultimately, po­­p­­ular opposition to the manner of implementation makes a worthwhile idea seem worthless as well.

Meanwhile, the reversal of local governments happened since while they were popular with citizens and local political workers, it was highly straightforward to paint them as the illegitimate engineering attempts of a dictator. Under this reasoning, provincial political elites were able to cement their own type of governing arrangement and recreate centralised rule in the provinces.

For context, a similar risk is posed if large-scale administrative change is pushed both without purpose-led clarity and in the face of popular opposition. Fiscally empowered local governments under the provinces, with reasonable judicial or even constitutional protections, might be the most appropriate concession at this stage. And one that may prove to be more sustainable even after the incumbent regime exits. The writer teaches politics and sociology at Lums. Published in Dawn, August 31st, 2026.

Umair Javed teaches politics and sociology at Lums. Find him on Twitter @umairjav.

For now, new units remains the part of the story worth watching, and further updates are likely as more details are confirmed.