Further to my post on debt in South Asia from November last year, I have been thinking more deeply about potential research ideas. I have returned to the “sovereign bank nexus” (SBN) and in particular how it manifests itself in Pakistan.

Why Pakistan? For one, it is a country that I work on regularly and am relatively familiar with. The State Bank of Pakistan (SBP) also has exemplary data availability since the country embarked on wide-ranging financial sector reform in the late 1990s. This allows for relatively easy data extraction.

Most importantly, Pakistan is a global outlier in the SBN debate: government securities account for ca. half of the banking sector’s assets, vs. an emerging market average of ca. 17%. Only Egypt and Ghana come or came close. We can therefore study an almost idealized version of the SBN here. This of course raises the question whether it’s more widely applicable.

A little more than half of Pakistan’s government debt is local debt. Given high domestic interest rates and short average duration of domestic government securities, the bulk of debt servicing accrues on domestic debt, about two thirds of which is held by the banking sector.

Pakistan’s structural fiscal deficits have seen the government become dependent on the banking sector to buy up newly-issued government debt. COVID-19 and Russia’s invasion of Ukraine have exacerbated this situation by exerting pressure on both revenue and expenditure sides of the budget.

Interest rates have risen in tandem with inflation. Banks hold increasing proportions of short-term floating rate sovereign debt and have thus become extraordinarily profitable, while pushing interest rate risk on to the sovereign. The SBN crowds out private sector lending and Pakistan has one of the lowest gross capital formation rates in the emerging world.

All the above is an important snapshot of where Pakistan finds itself at the moment. It might act as a good introduction to a paper that in its main part delves a little deeper into the SBN, in particular its origins, and potential remedies beyond the typical policy prescriptions.

IFI research has been conspicuously quiet on the origins of the SBN. However, reading a little more widely, a Deutsche Bundesbank paper found that the SBN increased in importance in the Eurozone’s periphery as capital requirements became stricter following the GFC. Whether or not this is applicable to emerging markets like Pakistan is unclear.

What seems clearer to me is placing the SBN in Pakistan within the context of financial sector reform. Natalya Naqvi of the LSE has written on this here and here, and this line of reasoning might be the most fruitful point of departure also for me. Can we link the growth of the SBN to liberalization and Basel adoption?

An orthodox line of reasoning might be that overambitious Basel adoption outpaced fiscal reforms. We thus dealt with a liberalized interest rate environment but unreformed government finances with structural fiscal deficits. The more government debt there is, the higher interest rates become, and the more incentivized to hold government debt banks become. Any favorite IFI topic would fit neatly in this narrative, e.g., SOE reform, tax administration, fiscal rules, etc.

A more critical perspective could be that financial sector reform intensified the misallocation of resources from long-term productive investment into non-productive activities. It led to a channeling of customer deposits into short-term debt, often for speculative purposes. Following the GFC, banks increasingly lent to the sovereign and the share of government bonds in total assets steadily increased.

In the absence of a quantitative model, can this paper’s empirical and narrative-based research approach provide new insights into the SBN in general and in Pakistan?

• The SBN has political economy roots; it is not apolitical and needs to be discussed in a wider forum than is currently the case
• The SBN is a consequence of market liberalization and Basel adoption in an “unsuitable” environment; and a degree of “financial repression” might be what is needed in emerging markets such as Pakistan
• This is especially the case in Pakistan where extremely low gross capital formation is the flip side of the coin to the SBN

For methodological robustness, I might need a comparative framework. There are several countries in a similar situation than Pakistan, e.g., Sri Lanka and Ghana–in both of which a domestic debt restructuring has recently taken place. Alternatively, I could compare Pakistan with other South Asian countries, and see whether Bangladesh and Nepal, with their directed lending, are some sort of counterfactuals.

Finally, it is interesting that Basel Committee on Banking Supervision is moving away from zero-risk-weighted government securities, and future accords might tie risk weights to external sovereign ratings. Until now it is in the discretion of the national regulator to assign risk weights to domestic government debt. Unsurprisingly, most if not all governments, including Pakistan’s set the weight at zero.

Given the strength of the SBN in Pakistan, the country is woefully unprepared for such a scenario, especially as its external debt is rated at junk. What is more, the SBP has recently suspended its only targeted regulatory tool to limit holdings of government debt (i.e., the “advance to deposit ratio”, or ADR, <50%), as the sovereign requires banks to finance consistent fiscal deficits. Whether or not this limit will be reinstated remains to be seen.