The perils of excessive specialisation and outsourcing - infrastructure and insurance

I blogged here examining the claims of Infrastructure Investment Trusts (InVits). 

At a basic level, an InVits is an example of unbundling and specialisation in the financing and management of infrastructure projects. There is a project sponsor or promoter who constitutes the InVit; a fund manager who is contracted out the responsibility of managing the investment; a project manager who is entrusted the execution of the project (say, construction or operation and maintenance); and a neutral referee or Trustee. What does this entail in terms of successful management of the scheme?

One, the project has to generate enough revenues to support atleast three purely commercial business lines (sponsor, fund manager, and project manager). Second, given the numerous conflicts of interests (related party transactions, revolving door of personnel etc) and incentive distortions (arising from holding durations etc), the Trustee/Board has to both have the expertise to monitor them and also the commitment to adhere to exceptional standards of governance in resolving issues as required. 

We know that infrastructure is low return asset. These commercial returns have to be squeezed out from this asset. The inevitable consequence is, as is well documented, asset-stripping and renegotiations, with all their public losses and controversies (see this). As to governance expectations from the Trustee, the stakes associated and the revolving door of personnel across different categories of intermediaries makes it perhaps too unrealistic.  

Take another example, of a national health insurance scheme that targets predominantly those outside the formal employment. There is a similar debate about the relative merits of a pure insurance and a public Trust-based insurance system. Like with InVits, an insurance system too has its set of intermediaries. There is a third party authoriser who contracted the pre-authorisations for insurance claims, the sponsor who issues the insurance, and the fund manager who manages the investments for the sponsor. 

Any health insurance scheme runs on three levels of diversification - across patients, across medical conditions, and across time. As I blogged earlier here, when the risk pool is mostly homogenous and consists of high-risk population category (poor), there are limits to such diversification. It is no surprise that such schemes have claims ratio which is pretty much close to 100%. And this is not to speak of the large subsidy in the premiums themselves. 

Therefore, the relevance of the traditional models of specialised commercial insurance become questionable. Neither are the margins enough to sustain them, nor is there likely to be a large enough portfolio of long-term capital that would need to be managed. 

Instead, a more likely effective approach would be a public Trust, which is professionally administered and is managed through appropriately designed service contract(s). Given that there is no getting away from public management of multiple contracts, it is a question of figuring out the the most prudent contract design that is consistent with the messy and uncertain practical realities of implementation. Fundamentally, the issue is about the state capacity to manage contracts. 

At a conceptual level, the limits to the Coasean bargain starts to become evident. The efficiency gains from splitting up activities and outsourcing them runs into both the costs associated with sustaining those individual business lines and the problems of managing the associated multiple contracts. 

It is important to keep this in mind as public policy encounters such financial engineering innovations in public contracting. Most such innovations, while irresistible in their logical appeal, have no track record of definitive success and endures just as is the case with such engineering in the regular financial markets. Unlike in case of the latter where private individuals are making choices with their money, in the case of infrastructure and subsidised health insurance, governments are putting tax payer money at risk. That requires a different level of scrutiny and standard for adoption. 

Postscript: 

Ananth sends me these comments on more reasons:

• Too many specialised roles make for dilution of responsibilities and only complicate simplicity
• What is the skin in the game for these financial entities? 
• Where is their expertise in managing infrastructure assets?
• Most importantly, what exactly is the underlying problem that is being solved or the gap being addressed? If there is none, why do we need a fix? - this might only make funding available for the limited category of financially viable assets. For majority of assets, where funding is needed and where cash flows can be turned over to the investors, the funding won't be available.