Estimating Loss Distribution for a Securitisation Transaction

In the latest edition of The Securitisation & Structured Finance Handbook 2018 (published by Capital Markets Intelligence) Vishal Saxena and Dilip Mohan from IFMR Capital have authored a chapter as part of the publication. The chapter discusses an approach to estimate the loss distribution for a loan portfolio. This loss distribution can be used to calculate the expected loss in an securitisation transaction, loan loss reserves, economic capital and value-at-risk. The authors first derive the limiting distribution of the portfolio loss as presented in papers by Vasicek (1987 & 1991) and then describe how they have extended the model to take care of the non-homogeneous subgroups in the portfolio.

Click here to download the paper.


The Nature of Financial Advice for Low-income Households

By Bindu Ananth

I was at an excellent behavioural finance conference organised by the Michigan University’s Centre on Finance, Law & Policy last week. One of the panels on investor protection debated issues including the impacts of disclosures, choice architecture and social norms marketing on investor behaviour. There was also an interesting discussion on role of advice and advisors in de-biasing investors or exacerbating weaknesses.

In the audience Q & A, in response to a question on the role of financial advice for low-income investors, one of the panelists responded that failures in the market for advice were less of an issue here since by and large, the right answer in most cases is just “save more for the future”. I found myself disagreeing with this notion strongly and one more reminder that the field of household finance has failed to examine the financial lives of low-income families in sufficient detail. In this post, I attempt to share from our KGFS work what are some of the other important aspects where advice seems to matter.

One, given that human capital (NPV of net lifetime earnings) dominates financial capital (wealth) for a low-income household, all of the issues around protecting that human capital is critical because that might make the difference between bankruptcy & resilience in the face of illness/accident/death. Most advice tends to focus on investments and the portfolio allocation question and surprisingly, pays little attention to insurance. Ibbotson et al (2007) provide a comprehensive framework to understand how human capital interacts with investment and insurance decisions. With limited resources, which members of the household should buy insurance? How much insurance should you buy? We find these are important aspects where households benefit from good advice. Specifically, insuring young, adult members of the household for the full value of their human capital is an important step. (One dilemma we faced was that a significant investment in increasing human capital that is made by households is higher education for children. The return to this investment depends greatly on the specific program and employability potential. We did not have the expertise to advise clients on this aspect but it feels like an area closely linked to the role of a financial advisor in this context)

Second, low-income households are typically saving and borrowing simultaneously despite a significant wedge between lending and savings rates (upwards of 20% most times). We don’t understand very well the determinants of this behaviour. Clearly, it is not always the right answer to save. High rates of return on micro-enterprises have been documented by Christopher Woodruff and others. Often it makes sense for households, particularly with surplus labour to borrow to put together the initial capital required to undertake such enterprises. Similarly, households with low but stable cash-flows (the village municipal worker for instance) may find it reasonable to borrow to build a house rather than wait to save up for the same. Working with the household to determine when to borrow and when to save and even combination strategies such as save for the down payment or borrow to save strategies could be very valuable interventions.

Third, the balance sheet of a low-income household has a combination of physical and financial assets. Physical assets such as land and gold dominate. On the liabilities side, there is a combination of formal and informal loans of different maturities. It requires serious skill to arrive at the APR of some informal loans! Which loans to refinance now that advances in financial inclusion are making formal credit more accessible? Which assets may be “dud assets” (ex: a piece of land that is not being cultivated) that could be sold to bring down debt burden? Which loans have a repayment structure that adds to the financial stress of the household? Working with the household to arrive at this comprehensive “balance sheet view” seems like an important role of an advisor.

Of course, there are significant challenges in converting advice into action and requires more careful work and business model experimentation. Equally, careful research and creating the building blocks for good advice for low-income households is also necessary and cannot be extensions of existing advice frameworks. The myth that these households have simple problems that require simple fixes & simple products needs to be challenged by researchers and pioneering providers.


Stress Testing Methodology – Brief Comparison Across Regulators

By Nishanth K & Madhu Srinivas, IFMR Finance Foundation

The below table summarises, along some key dimensions, the stress testing methodologies adopted by the central banks in India, US, UK and EU to assess the stability of their banking system. It is to be noted here that the stress tests that individual banks conduct by themselves, as part of their Internal Capital Adequacy and Assessment Process (ICAAP), do not figure in our comparison. Also the below analysis is based on the stability/stress test reports of the respective regulators for the year 2016.

All data for the above comparison was taken from the following references:

Click here for PDF of the infographic.


Natural Catastrophe Insurance – In Conversation with Mr. Ulrich Hess

By Vipul Sekhsaria, IFMR Holdings

In the below video we share a brief conversation with Mr. Ulrich Hess, GIZ. Mr. Hess is currently a Senior Advisor, InsuResilience Initiative at GIZ, and has worked extensively in the field of natural catastrophe risk insurance market. In the video he shares his insights on the impact of natural disasters on the livelihoods of households and the risks associated with it. He also talks about the challenges in designing a natural catastrophe insurance product and addressing issues associated with both inefficiencies and effective delivery of the product.


Developing the Natural Catastrophe Risk Insurance Market for Low-Income Households in India

By Vipul Sekhsaria, IFMR Holdings

Natural disasters leave behind them a tale of death and destruction that affects the economy on the whole and severely impacts communities, especially low-income households, which bear its brunt. While little can be done to prevent natural calamities like floods, cyclones, drought etc. from occurring, what perhaps can and should be done is how best households, especially the vulnerable ones, can mitigate the financial losses that such calamities have on their lives.

Flood & Drought Risk

In terms of number of people affected, India tops the list of 163 nations affected by river floods as cited by World Resources Institute[1]. Close to 76% of India’s 7,516 km long coastline, is prone to cyclones with over 40 million hectares (12 per cent of land)[2] being prone to floods and river erosion. Floods can severely disrupt livelihoods, especially in low-resource settings. Flooded households are affected by a plethora of adverse conditions including food insecurity due to crop failure or affordability concerns due to sudden price changes. Daily care of children is importantly challenged during floods as in worst scenarios all basic services become disrupted, including water and sanitation conditions, or the provision of basic community health and social services.

Like flood, drought in India is also a major disruptor of financial well-being with 68% of the country being prone to it in varying degrees[3]. It is difficult to provide a precise and universally accepted definition of drought due to its varying characteristics and impact across different regions such as rainfall patterns, human response and resilience etc. Last year (2016) more than 300 million people living in 256 districts were affected by drought after two years of sparse monsoon rains[4]. The latest findings suggest that while there have been alternate dry and wet spells over the past three decades, the frequency and intensity of drought years has been increasing – for instance Tamil Nadu was declared drought hit in January 2017 after it recorded the worst rainfall in 140 years[5]. What’s important to note is that while the direct effect of drought could be on the farmer and the agriculture economy, but due to its high incidence, the local rural economy also gets severely affected thereby expanding its impact base beyond the farm sector to rural labourers and small rural businesses.

Natural Catastrophe (Nat-Cat) Insurance

Given the fragile economic livelihoods of the underlying households that microfinance institutions and small business lenders serve, even significantly diversified originators typically have a large percentage of their capital at risk in case of a localised natural catastrophe, resulting in a higher cost of capital. This leads to either no catastrophe cover or cover that is unaffordable to people living on low incomes. Further a majority of households never have access to any insurance that protect their assets and livelihoods in the event of a shock. The existing PMFBY (Pradhan Mantri Fasal Bima Yojana – Prime Minister’s Crop Insurance Program) is a restructured Weather Based Crop Insurance Scheme covering only Farmers – it does not take care of many other rural customer segments like Labourers, Small businesses that form 60% of the rural population. Even for farmers it doesn’t provide the much-needed liquidity during the constrained circumstances of a natural disaster like flood nor any protection towards assets other than crops (example: house & contents, livestock, other small holdings). The PMFBY structure is also highly subsidised by the government (to an extent of 90% subsidy)[6], which is a good first step to drive adoption, but without an exit strategy, the long term continuance of subsidy always remain questionable.

India was the first developing country to pilot weather indexed insurance and, despite the recent spread of weather indexed insurance programs across the world, more farmers purchase weather indexed insurance in India than in any other country. However, despite the large public subsidy, as mentioned above, a significant majority of India’s farmers have remained uninsured largely due to issues in design, particularly the long delays in claims settlement.

In terms of product development, designing an Index Based Parametric Cover is somewhat comfortable at a portfolio level rather than at the individual level (micro level), since at a portfolio level, rate makers have access to more managed data of the spread and concentration of assets across the geography. The return periods of the calamities and the portfolio data make it possible to arrive at a commercial rate for the Index Based cover. Recent experience suggest that while products are available but they are also limited to perils like Earthquake which are usually perceived as low-frequency event affecting a much smaller geography in India and therefore are of lesser demand as against for Flood and Drought.

More products for protection around Flood and Drought should also appear in the near future but cost of such solutions is yet to be evaluated. It’s worth mentioning here that, trigger of such portfolio level product results in a payoff to the risk originator (Micro Finance Institutions or similar) to cushion their own portfolio from delayed receipts of the loan repayments due to the stressed situation caused by the catastrophe. The challenge in this segment as it seems is that most originators who are already working on tight margins find it difficult to cover the cost of an earthquake protection product at a portfolio level and the high price still continues to be a dampener.

Designing a Nat-Cat Micro product

While the subject of Index Based Parametric cover is largely centred around loss of assets (whether fixed or movable), there has been very little or no work done so far as to protect the loss of Individual Income due to the incidence of perils like say, flood and drought, through an Index Based Parametric cover. The advantage of originating such cover is making the end consumer (micro level) ‘Nat-Cat-Resilient’.

The biggest challenge in developing the Nat-Cat Micro product is the absence of structured income data at the micro level. In absence of any close estimate of the different income profiles and the effect of Nat-Cat perils on this income, it is not possible to initiate the ratemaking of the risk – ‘Loss of Income’. Since the potential customers are mostly from unorganized sector, a great deal of primary research work will be involved in estimating the different income profiles of the constituent occupation classes.

To address this challenge we have undertaken a detailed primary research activity (details on which we will share in subsequent posts) to capture insights on the impact of natural calamities on income of rural customers, length of the impact as well as coping mechanisms. Joining in this detailed research work is a leading DFI (GIZ InsuResilience Direct Insurance Implementation Team) who has partnered with IFMR Holdings (IFMRH) in developing Catastrophe risk protection market along with weather based technical service provider based in India. In its current phase the goal of this project is to develop probability curves that can be externally assessed and then used to pilot differing approaches like the one detailed above as a “Micro Nat-Cat Product”. If successful, the aim would be to make these probability curves available to others to develop similar coverage and products to serve a much larger population in India.

As part of this blog series we intend to share insights from our research and interactions with expert stakeholders in subsequent posts.

[1] http://www.livemint.com/Politics/hjUVTrwyI0I4p4b4enBg1K/India-tops-list-of-nations-at-risk-from-floods.html
[2] http://www.worldfocus.in/magazine/disaster-management-in-india/
[3] http://www.ijesmjournal.com/issues%20PDF%20file/Archive-2017/Jan-Mar.-2017/4.pdf
[4] http://www.thehindu.com/todays-paper/tp-in-school/Reeling-under-dry-spell/article17052569.ece
[5] http://www.business-standard.com/article/economy-policy/ne-monsoon-worst-in-140-years-144-farmers-dead-tn-declares-drought-117011100782_1.html
[6] http://indianexpress.com/article/business/business-others/pradhan-mantri-fasal-bima-yojana-crop-insurance-plan-to-entail-rs-8-8k-cr-outgo/