Before we dive in it’s worth pointing out that it’s hard to be generic about something that doesn’t exist :-)
Not just is P2P a poor term to describe the online lending/borrowing Fintechs, it is a very heterogeneous sector.
Or to simplify a bifurcated sector.
On the one hand we have the “you-select” model. You go to a platform and you pick the asset you buy.
On the other hand we have the “they-select” model where you go to a platform and click on a rate and they pick the assets you will buy.
The former is more akin to a market – you go there and you buy.
The latter is more akin to buying into a fund managed by a fund manager.
Very different animals.
OK lets brush all that to one side and dive into eight areas of potential Risk Debt in P2P right now. I’ll not fret too much about who will, in extremis, end up owning the risk – lender, investor, or management – after all we rather hope that all their interests are perfectly aligned don’t we?
The tide always comes in and always goes out (or with the soon-to-be ISA the tsunami comes in once a year).
And as King Canute showed there ain’t nothing you can do about that no matter who you are.
All platforms rely on there being an incoming tide for anyone that needs to exit (and they do disclose this although some websites are now so large it would take a long time to read it all).
One way round you might invest for a long term and get caught short if having stuck your money somewhere for five years, in three years the tide has changed and plenty of folks can’t get their money out when they need it.
The other way round you might invest in an apparently short-term product but get locked in for longer if the underlying assets are longer (“maturity-transformation”) and no inflow is available to rescue you if you need your money back after the apparent term.
To FS folk liquidity is everything. Liquidity problems always lead to reputational blows to the sector.
As #oldFS knows a reputation for trust that has taken decades to build can be wiped out in weeks.
The 2008 crisis was started by an ever tighter and less liquid interbank market. Banking in mediaeval times only really took off (in this debt-fuelled way) when the interbank market (providing … liquidity) came along.
I rather feel there is an opportunity for innovation here – liquidity is everything in a world always in flux. It’s tied up with secondary markets – if I can’t sell my asset to the originator – how many cents on the dollar will someone else pay me for it?
2) Principal Risk vs Agent Risk
P2P passes on principal risk from borrower to lender.
However even though you end up with a principal exposure to the specific asset (/portfolio) you also have an agency exposure to the platform.
The platform needs to source quality deals (especially if you ticked the “reinvest income” box), filter out fraud, service payments flows, chase late payments, work out defaults etc etc etc.
Even if technically your asset exposure doesn’t change if a platform disappears your asset servicing (“agency”) exposure certainly does. And getting all the information to a new agent (or even finding one in some cases) is definitely non-trivial. Not least of which you yourself have limited information on your asset (for consumer lending you don’t even know the name of the person you lent to).
How well firms plan to pass on the asset-servicing in the case of wind-up varies widely – and, as we will come onto in Legal Risk, how well it will work in practice remains to be seen.
Furthermore the division between principal and agency risk isn’t always as black and white as it sounds.
I have however seen various rate guarantees or tracker rates that look to me like they might be potentially, at the margin, introducing an element of principal risk into the equation.
Another FS-y point is that agency risk can flip into principal risk in extreme circumstances (which happen all too often in FS). In the recent Swiss Franc crisis several “agents” went bust as agency risk flipped into principal risk.
Whether just agency risk, or a little bit of principal risk, or in extremis a “flip”, in general the risk of platform failure (and hence your risk on a platform) is impossible to ascertain.
In general platforms provide little financial information that would enable you to, as it were, credit rate them.
3) Provision Funds and the Art of Self-Insurance
The “they-select” model can often come with a Provision Fund which aims to stabilise returns. As I stated in “How To Quantify Lender Risk In P2P” this May Well Be A Good Thing, and Have Worked Well So Far but it certainly complicates any quantification of the risk.
More subtly I have been amazed that next to no-one I have spoken to has noted that there are (at least) two models of Provision Fund.
One type of Provision fund (eg Zopa), which we may call a “individual” provision fund, pays out on individual impaired assets.
In the case that it runs out of money it says “sorry” and no longer pays the owners of the impaired assets.
The other type (eg Ratesetter), which we might call a “communal” provision fund, also pays out on impaired assets.
In the case that it runs out of money it says “sorry” and all the assets are then (de facto) pooled. In essence this spreads the investment risk across the portfolio.
In the first case as a lender you are exposed to your assets that were allocated to you.
In the latter you are exposed to the whole portfolio – in a way the asset ownership can end up being a bit nominal.
More broadly the topic of provision funds and insurance and self-insurance can fill whole conferences. It’s complex.
It’s very hard to assess the risk from the outside. It’s especially hard to assess the chance of something which has succeeded for n days in a row (where n is “many many”) failing “tomorrow”. That’s why insurance firms have actuaries.
As a business risk if firms grow and grow and have conservative provision funds this in itself can attract vultures who want to buy the firm and strip down the provision fund.
In an un-actuary-ed world there is is significant scope for managing provision contribution rates (cf pension fund contribution holidays). As we shall discuss below re “yield-driven marketing” this directly affects headline rates (and hence asset growth) so there is a conflict of interest there.
It will be interesting to see how this whole P2P-provision fund space evolves. Plenty of scope for it to evolve well or “not so well”.
4) Yield-Driven Marketing – Ooh, Ouch
As punters are used to percentages for deposits which are, well, actual numbers, they aren’t at all used to estimates/guesstimates. Nor have I seen this discussed much.
In a world (and comparison sites, and advertising) driven by headline numbers, the “Big Figure”/“Projected Return” on platforms is generally a complex model something like current yield minus fees minus projected losses [plus payments from provision fund].
That’s a model.
Furthermore as I mention in “Do P2Ps Communicate Lenders Risk Well?" we can see that platforms interpret the most basic P2PFA terms completely differently:
“Looking at the key “expected losses” – some (conservative) platforms seem to treat that as “expected maximum losses” and some as nearer the mean of expectations.”
There is clearly a lot of leeway for platforms to end up with different numbers even if they had the same portfolio.
And to put that in context, in grown-up FS, Basel is rethinking the whole idea of banking internal models for capital adequacy purposes as internal models have differed so much [see https://finsia.com/news/newsarticle/2014/11/17/basel-questions-future-of-bank-internal-models including interesting comments on low risk loan risk calcs].
One of the side-effects of becoming “more FS” is that P2P will gradually get more absorbed into the world of FS-type regulation.
That’s a world where, especially Anglo-Saxon, regulators love fining firms tens of billions for simple words like mis-selling (as defined after the event – never before of course) which also launch a zillion spam phonecalls.
We don’t want be receiving spam calls in a few years time about P2P “mis-selling” do we?
What do you foresee?
Much talked about not so much practiced?
My favourite statistic is that last autumn Lending Club reduced the number of data points it provided to lenders by 44. AFAIK this number of data points is more than UK platforms generally provide in the first place.
It is interesting how often one doesn’t even get informed of the rate the borrower is paying. You’d like to know – right?
If you can’t see the list of all the assets it all becomes a bit black-box like doesn’t it?
A bit like a marketplace where everything is under the counter and unseen until you buy it…
6) Cherry Picking
Whenever I think about the above point re very little data I always wonder how the funds that invest in P2P assets manage to add all the value they, like all fund managers, claim to add in return for their fees.
Do retail investors choose assets at the same time as institutions?
Or do institutions get first bite at the cherry?
Do institutions get superior information to Joe Public?
Are there standards across the industry or does your favourite platform treat the institutional investor better or pari passu with the punters? I think we know it won’t treat them worse ;-)
7) Legal Debt
Many of the above issues have legal angles.
The law is a funny thing. Makes a lot of money for lawyers so they don’t mind.
But it’s only when something has been tried and tested in the courts that you know whether that expensive document you have was worth what you paid for it.
Right now P2P has a whole variety of highly important structures. And from what I hear there is rather more than a splash of “Legal Debt” in P2P.
For example around provision funds – boards, trusts, arms-length, semi-arms-length, connected. And what about investment guidelines? (an absolutely non-trivial issue I can assure you).
For example around platform collapse. Having seen firms collapse things like “who has the key to the safe”, “the passwords” etc etc become of immense practical importance to (as they do in disaster recovery – which can be all too closely allied to legal risk).
This is probably one of the loosest nuts and bolts on the whole P2P show.
Partly because it hasn’t had many test cases.
But also as, if one is a startup that grows fast, it’s one of those many lags – tech debt, risk debt, legal debt, etc – that end up needing paying off sooner or later.
One way or another.
8) Cyber Risk, Fraud et al
Talking of disasters…
Cyber crime is of course a huge issue affecting all businesses right now. And all businesses do something about it (we hope).
But if the Chairman of HSBC says its one of the major risks a bank, of their size, with their resources, faces then you can be sure it’s a challenge for anyone.
If you haven’t read it I recommend “Anthony Hilton: Government must help fight the cyber threat”.
As is fraud (Citigroup recently got stung with a biggie – and again they have a lot of resources).
Would you be really surprised if you read tomorrow that a platform has had a huge hit in this area?
I can think of one P2P (but not any/many more?) that insures against this risk.
That makes a lot of sense to me as this kind of risk can only be borne with huge market capacity. I personally know FS firms – from scaleups to listed - that have been sunk or holed below the waterline by fraud let alone the new frontier of cyber.
If You Don’t Pull All Risk Debt Into One Place It Will Slip Through The Cracks
Enough to keep y’all going?
Which of these do you think is key?
Which biggies do you think I have missed out?
To recap a point I made in the parent article… One or two folks wondered what “other risks” there are other than credit and compliance.
Well I have discussed eight above but the list is limitless – businesses can go wrong in so many ways - operational, reputational, performance, strategy, business model, audit, growth (too fast/too slow), disaster recovery, fraud, cyber-crime, etc etc etc and all those items that fall between these “square floor tiles on an uneven floor”.
P2Ps need a process for ongoing capture, tracking, and management of all risks that might derail them and their lenders/investors.
Having been in FS before CROs existed I have lived and breathed a world where there is no-one whose sole task it is to keep track of everything firmwide that can derail the firm.
De facto the CEO does that. But in FS there are so many moving parts he doesn’t really have the time along with all the many other responsibilities.
If we are to avoid “Deja-Vu All Over Again” P2Ps should learn from the mistakes of #oldFS and get a (proper – ie businessman not box-ticking) CRO before the horse is well down the lane with a broken leg needing to be shot.