When the firm filed for Chapter 11 on September 15, 2008, it had ‘assets’ valued at just shy of 700 billion US dollars on its balance sheet.
While I was lucky enough that MAS ringfenced certain assets of the Singapore entity of Lehman to continue paying salaries and Nomura’s acquisition of personnel happened quite quickly, the double-edged sword of leverage has continued to fascinate me.
Fast forward to my present. Wearing a VC hat now, there’s one question that has consistently cropped up during my conversations with several founders running non-lending businesses ranging from ecommerce platforms, supply chain SaaS software players, ride hailing platforms (the list is endless): “Should I get into lending myself?”.
To come up with an answer is very difficult – and, hence, the article – but the rationale of why the question is even being asked is relatively straightforward. To turbocharge growth. Lending to one’s customer base, especially in hyper competitive verticals like ride hailing/food delivery, can enable a business to achieve positive unit economics while boosting top line. Lending via one’s own balance sheet enables the startup to achieve the same but on steroids as you control the spigot of extending even more credit yourself all the while ensuring a more seamless customer journey. But, in the long run, this seemingly innocuous question opens a Pandora’s box of possible future scenarios, ranging from an eternal money grail to – sadly, in many cases – liquidation. And this is because, in and of itself, the path traversed to get the business to this point, doesn’t necessarily provide the foundation to evolve into a successful lender.
Prior to embarking on using ‘one’s balance sheet to lend,’ evaluate all other possibilities too like partnering up with Embedded FinTech startups and see which option makes the most strategic sense given the finite resources in-house. But IF, the decision has been made to go down this path, broadly speaking, I feel there are 6 main points a founder must consider:
1. Proprietary data development
One of the main differentiators in a sustainable, profitable at scale lending business is to curate a proprietary dataset that yields the required insights into analyzing the credit worthiness of a borrower. The lending fintech ecosystem graveyard is littered with several startups that in spite of raising tremendous amounts of capital (both equity as well as debt) have severely mispriced credit risk of the borrowing entity – be it business or consumer precisely for this reason. While there might be tremendous inefficiencies and traditional tech stacks powering many banks/NBFC’s credit underwriting is an extremely difficult skill set to curate and continuously fine tune over time.
A key differentiator/competitive moat that any lender should continuously try and expand on is the ability to mitigate credit risk by generating or extracting from a variety of data sources a deeper understanding and profiling of the borrower.
This can take the shape or form via:
- Hardware integration: Example – POS front end enabling providing real time transactional data.
- Software (Direct): Powering business’s erp system, spending/saving apps, credit card info aggregator, provider of ecommerce/travel services.
- Software (Indirect): API connectivity to numerous government databases/credit bureaus/alternative data sources including type of phone and location tracker etc.
Being able to meet the massive underserved segment desperate for credit in a seamless and quick turnaround time and utilizing a proprietary credit underwriting method will enable a FinTech lending startup to scale up quickly without taking undue risks on their balance sheet.
2. Source of capital/assets liabilities matching
FinTech startups typically fall into a trap when trying to manage their net interest margin (NIM). High cost of funds forces them to go further down the credit curve and extend lines to the very high yield segment to make the business model work. Some larger FinTech players in the later stages, try to secure bank licenses to minimize their cost of funds by becoming a licensed deposit taking institution despite increased regulatory scrutiny, which may be minimized by adopting a tech first approach to all facets of the business. Combination of lack of track record coupled with the amount of capital to raise is the conundrum that the founder needs to resolve as quickly as possible.
The path typically taken by a new startup is to:
- Equity on their balance sheet (the most expensive, but typically the only one they have access to)
- Raising venture debt typically 8-18% p.a. excluding 10-20% of warrants attached
- Credit line via more established financial institutions (cost ranging from 4-10% depending on which segment of lending one is targeting).
Whenever debt is involved, pay special heed to the amount of leverage you are taking on along with the covenants mentioned in the term sheet that might be quite onerous on a startup especially when the rosy future plan doesn’t pan out.
Another factor founders should watch out for is ensuring the asset liabilities mismatch is minimized i.e. tenor of capital borrowed vs capital being lent out has enough safety buffer. Ideally with clauses of extension of credit limits and rolling over of debt contingent to staying within the required covenants.
3. People and team structure
While it is advantageous to look at how larger financial institutions are organized, startups do not have nearly the equivalent amount of resources at hand to achieve the same. The overall team structure is an integral part of being able to achieve long term success.
It is of utmost importance to ensure that there are:
- Clear reporting lines. For instance, the risk team in place is completely independent from the sales department with direct reporting lines to the Founder. This is to ensure that undue pressure in terms of ‘scaling up growth taking upon any risk’ mentality is not prevalent within the department. Healthy friction between the sales and risk departments leads to building a sustainable business.
- Ensure clear and granular KPI’s are provided for each department from front origination to risk to the tail-end of collections.
Depending on the go-to market (GTM) strategy of the startup, there might be a treasure trove of data, but the risk/credit department and functionality along with robust processes would not have been created. A lending marketplace, for instance, where the sales team may have the green light to bring any and all borrowers on to the platform with the credit underwriting requirements borne by the actual lender.
When looking for a risk officer, one may need to be cautious of having a highly credentialed risk officer from an established financial institution whose method of doing risk analysis profiling might suit that of a bank but not of a startup looking to disrupt the space.
4. Credit control, scoring and NPL process
The key to a streamlined and smooth functioning lending setup lies in having:
- Clear process and documentation of understanding and identifying ideal profile of borrowers
- Credit scoring/risk analyzing and last but not the least
- Action plan of when loans do roll over to delinquent and non performing buckets
Regardless of how good one’s data set/AI models for risk analysis might be, NPL is an issue that no lender can overlook. Identify and lock down in advance what the process is to handle maximization of recoveries of such cases by considering the following:
- When should pre due reminders be sent and through what form?
- Is the origination team responsible for the above (potentially have better relationships with the borrower) or to implement a third party tech solution or to outsource to digital credit management companies?
- Till what bucket of date past due (DPD) will the cases be managed in house or outsourced?
- In the case of the former, the size and scale and hence budget of the inhouse collections department relative to the size of the loan book coupled with developing an in house technology solution or sourcing the software from an external one?
- If the latter, onboard how many collection agency vendors by creating a clear evaluation checklist along with data security checks is vital?
- If multiple agency vendors, what is the time allocation of cases to each along with the process of rotation of cases between them?
- When should the NPL be sold off the balance sheet?
Given the lack of reputation from the collections industry some additional points to consider prior to dealing with them.
- Ethical standards of collection process – does it involve field collection, if so what checks and balances are present?
- Any relationship with someone in the lending team? (as the industry is infamous for kickbacks)
- What is the quality control process when it deals with calls between the operators and the borrowers?
When dealing with NPL, the stage of the startup is highly relevant. Evolving from primarily outsourcing all cases, to slowly scaling up an in-house team to deal with cases up to 60-90 DPD and subsequently outsourcing it to 3-5 collection agencies, rotating cases between them every 1-3 months and aiming to sell off the bad debt after 360 days. This is something that Founders should always keep in the back of their minds while scaling up the startup.
When it comes to underwriting loans, the Company’s screening and evaluation process ensures that underwriting policies are properly developed, implemented and adhered to. Balancing the loan portfolio with low/high risk, tenure with terms of financing, allowing greater risk adjusted returns.
5. Performance and risk monitoring
To facilitate effective and efficient resource allocation at the Company level, tracking the right metrics is paramount. And feel free to keep them as granular as possible to ensure a transparent picture ranging from the team performance to quality of the loan book can be drawn.
In addition, a proper set of risk management framework allows the Company to monitor and respond to changes in credit risk over the lifetime of a customer in a systematic way. Some topics to consider:
- Establishing effective communication channels between the Company and borrowers for regular business and financial updates.
- Automated live updates of borrower's payment systems, accounting records, ecommerce accounts, advertising accounts, bank accounts, etc. Allowing real-time updates from the source.
- Internal tools that facilitate portfolio risk monitoring in a consolidated dashboard and machine learning systems to identify and flag deteriorating borrowers.
- Action plan for deteriorating borrowers.
6. FX Hedging/Mismatch
While forex is something that can’t be hedged at the relatively early stages of a startup, for Founders dealing in Asia, this is one aspect that can ruin the best run businesses.
Foreign exchange risk happens when the currency of capital being raised through debt is different from the currency that is being lent out. This may create a vulnerable position to the startup as it is highly susceptible to macro risks.
Domestic currency blows up during a crisis (Asian Financial crisis, GFC etc). Flight to quality typically takes place, leading to strengthening of the USD or EUR (currency of debt capital raised) against emerging market currencies (currency where loan is issued). If this happens, your asset/loan book will be less valuable as compared to debt that has been raised.
Ideally debt to finance the loans to be issued will be raised in the same currency but typically most institutional funds will issue debt in hard currencies - USD/EUR/JPY.
Couple of ways to mitigate this risk:
- Try and raise debt at the local level (across Asia there are multiple family offices/funds that would be happy to lend money in the local currency) or,
- Agree with the lender on a cross currency swap on its end prior to lending (some funds actually prefer this as it reduces their risk of the startup going belly up because of an unfavorable FX move though this increases cost of capital raised).
- Once the startup has achieved a certain scale, it can explore hedging the risk on their own. Example: Engaging in USD-XXX forwards/non-deliverable forwards (NDF) via a discount brokerage platform like Interactive Brokers or Saxo Bank.
In conclusion, I hope the above sheds some light on the various factors, external and internal, a founding team should consider prior to going down the path of becoming a lender off of one’s own balance sheet. Lastly, given the macroeconomic headwinds, partnering with an embedded FinTech player to create a test bed of financing options before going headfirst into providing the same could be a smart decision.
I am happy to have a more detailed discussion with any founder/founding teams who are looking to expand their core business with a lending angle. Reach out to me here on LinkedIn or write to me at email@example.com