BNPL: Some insights from a successful short seller in the space

Thanks very much for you time and any insights you can share Andrew. It’s always valuable to hear from veterans who actually do the work.

Have tried to keep this fairly succinct but please feel free to add anything that you feel would be useful to those of us with less experience than yourself.

1. If you could give us a brief background on your experience as an analyst, particularly with respect to financials and insurance companies it would be appreciated.

I was a sell side Australian banking analyst in the late 1980’s and early 1990’s as well as an insurance analyst at that time.  The timing is quite important since Australian banks went through a major bad debt crisis in the early 1990’s as the economy and asset prices imploded, washing out the impact of the 1986 introduction of foreign banks and a period of extravagant growth in lending.  As a consequence, even though the credit loss environment in local banking has been benign for quite some years, you never ever forget the lessons from when major institutions skirt with bankruptcy due to delinquent loans and inadequate provisioning in such an environment.

Understanding insurance is a real benefit in BNPL.  Both life and general insurance have essential elements of customer acquisition costs (CAC) versus lifetime value (LTV) of a customer.  Why do you think you can renegotiate your car insurance fairly easily – because the insurer wants to keep you rather than incur the cost of new customer acquisition.  In life insurance, the LTV of a customer is a specific actuarial calculation to evaluate profit and net worth of the life company, and is, of course, subject to significant shifts in a number of variables – mortality, interest rates etc.

2. To the best of your recollection, what were your first impressions of the BNPL sector after Afterpay first brought the sector into the spotlight post-ASX listing?

I was very cynical because all the bulls were telling me it wasn’t credit provision - when it patently WAS - and that the customer data was worth squillions.  Data has very limited value when it is of a group of folks who are the extreme end of the credit spectrum and have great difficulty repaying their obligations.  The first thing an analyst should look at with a credit provider is the credit loss experience, not growth in lending/business, because if the credit loss is consistently high, the business will not get funded on the liabilities side and more equity will be required, until the appetite for that subsides, and the enterprise defaults.  These credit loss metrics, when you assessed them versus the average stock of loans – accepting that the loans turned over 16x a year – were unbelievable, in the high teens percent per annum, versus banks at 60bp even in the hefty reserving of H1 2020 for COVID impacts.

The other cause for cynicism was the fact that around the globe, credit providers trade at very modest multiples of earnings, which partly reflects the balance sheet gearing.  Moreover, the higher risk the provider – think US credit card companies – the lower the earnings multiple, down into low single digits.  Accepting these were new companies, upon maturity (if they got there) why would they trade at growth stock multiples? Hence, on any long term view, investors would be in for aggressive multiple compression.

3. Why do you think it is that the market took to the idea of cheap and easy credit so willingly despite clear indications that the economics were questionable for many players?

In all seriousness, this is a sectoral story where Australia’s capital markets practitioners should hang their heads in shame.  The sheer greed, backslapping at discovering Afterpay when it was really just a fantastic market-play and willingness to finance appalling smaller companies for egregious fees at the expense of the retail bag-holder is a sorry tale.  It was driven by FOMO, some spectacular marketing, near fraudulent analysis by the sell-side.  I have rarely seen such garbage research on companies in decades – eg Ord Minnett valuing Sezzle at 22.5x revenues in a research report, being the SOLE valuation metric.  The appalling sell-side research reflected brokers’ desire to participate in the inevitable high fee equity raisings for the companies.  The immense marketing of the “story” in the media and creation of new superstar personalities, whose main strength was access to capital markets and the ability to heavily send the proceeds, was execrable.  UBS are to be commended for their proper analysis and subsequent cynicism.

4. We’re all familiar with your work with respect to SZL in particular. The stock has fallen over 90% from the highs of early 2021 at the time of writing. What attracted you to the stock vs others in the sector, and how would you rate their model vs peers? Did you make money on the short?

Where do I start??!! How long have you got!

 (1) It only operated (realistically) in USA – US credit standards, notably the acceptance of “non-recourse” are worse than Australia so I felt a product like this would be abused with high bad debts. That has turned out to be the case – in spades.

(2) Disclosure – once SZL started filing 10-Q after an S-1, we had a much better look under the hood of a fairly simple businesses than we would get at Afterpay or ZIP.  SZL was a Lada, not a Ferrari.

 (3) the funding structure was unsustainable with the use of high interest loans from tertiary credit market players (eg Bastion who also financed Zebit) at low teens interest rates which were always being refinanced (with re-fi fees!) then replaced with the “merchant interest program” whereby the merchants lent back their money to Sezzle at a few percent over LIBOR.  That turned them into a “bank” where the funding could well be subject to a run as the merchants could withdraw – with limits – effectively on demand.

(4) this was important because we were able to show the LIABILITY/FUNDING structure of the company was unique, which allied to operating losses and bad debts, was going to force more equity issues or bankruptcy.  This has effectively happened. The analysis of liability structure was unique in BNPL and nobody else bothered with it who claimed to own or publicly analyse Sezzle

(5) the insider ownership, whilst escrowed, was horrendous, with two years work, US$6.5million of losses and an US$11.8m investment – worth 13c per share -being sold to investors for $1.22/share (so say a 900% uplift) and ultimately being priced at 10x that IPO figure.  Management would obviously be sellers when they could – but by then, the game was up to a large degree.

(6) management presentations were farcical – bad debt barely rated a mention but it was putting the company out of business.  The CFO couldn’t (wouldn’t??) answer simple questions like how many merchants were in the merchant investment program which would give you a clue as to its sustainability.  There was never anything negative to be said.  

(7) with some help from fin-twit friends, we believed the company was doing very high risk business in hemp products and “interesting” (depending on your proclivities) anime products.

Sezzle’s model is awful versus the peers and I have no idea why ZIP are buying them, especially for nearly A$200million. On current trends, shown clearly in the March 2022 quarter 10-Q, in my opinion, without the deal, Sezzle will end up filing for Chapter 11 bankruptcy protection.  The equity market’s reaction to the ZIP deal suggests other people don’t get it either. We did make money on the short, but had to be more cautious than our analysis warranted due to the large insider ownership and the Chairman’s undoubted ability to engineer deals such as that with Discovery.

Sezzle provided two good war stories. Myself and @ignore89039500 did a twitter spaces call in late June 2021 when we attracted nearly 80 people to discuss the story - one of the first uses of that medium; the company got wind of this and described us as “comical” in a Sydney Morning Herald article (1 August 2021).  The second war story was from that same article where the Chair described Sezzle as being “essentially training wheels for credit”. Who the hell invests in a company that is “training wheels for credit”.  With a few words, the Chair and largest shareholder had proved our thesis for us!

Finally one of the biggest issues in the sector was finding the “tipping point” from where the direction of equity prices was down so that you were shorting a fundamental issue, not just a valuation.  All of the shares were overvalued in late 2020 but most doubled in the space of three months in early 2021, which would have blown you away without exposure limits.  Shorting this sector was hard and we had a few false starts before it paid off. 

5. Are there any players in the sector that you are familiar with that you believe have both the funding and business economics to survive (absent material changes in their operations / funding sources from those that exist today)?

Let’s do some simple economics using an annualised version of ZIP’s first half FY2022.  Transactional income is a ludicrously high 6.79% of underlying sales – that can be pressured by competition.  Net income before interchange fees is still over 6% but hefty fees take 1.25% of sales away.  Hence, net income from the business before overheads and credit losses is about 4.75% of underlying sales.  We might call that dollar income “variable” as it is margin and volume dependent.  But the aspect so many people forgot was that it was placed on top of a fixed cost business where costs were growing rapidly – people, IT and especially MARKETING.  In addition, share based payments were very high.  ZIP cost base ran at an annualised $500million in H1FY22, of which (annualised) $150million was marketing spend.  If that base is maintained and credit losses run at only 1.5% of sales (they were well over 3% in H1FY22), underlying annualised sales need to be over $15billion to break even versus H1FY22 annualised of $8.9billion.

In my opinion, the fundamental (negative) turning point in the sector started to come from two areas:

(a)    that bad debts rose sharply as stimulus payments in US and Australia were withdrawn – delinquencies have not yet really abated; and

(b)    the cost of acquiring each new customer has skyrocketed.  For example, on my analysis, the cost of acquisition PER CUSTOMER for Afterpay nearly tripled between June quarter 2019 ($10.79) and June quarter 2022 ($31.31)

It’s hard spending money on customers when 70% of your top line income goes in bad debt charges, and that’s what happened to ZIP in H1 FY22.  As a guide, when Westpac provisioned hard in early CY20 for COVID – not knowing the full impact – it lost only 22% of its banking net income in bad debts.

I think BNPL as a concept will survive but needs to be better managed.  It’s questionable in a standalone company, but has value in a diversified financial lender. That’s why CBA will make it work.  The obvious player is LFS having acquired Humm, as it knows how to operate in the higher risk personal credit/ merchant relationship environment.

6. What advice would you give to a BNPL CEO / CFO team that is struggling right now, as many of them are?

Stop and think.  The worst things CEO/CFO’s in BNPL do is to be growth obsessed and sprinkle shares around like confetti.  It’s just endless dilution of shareholders to merely survive.  Look at ZIP paying $200mn ish for a A$17m equity balance sheet that looks primed to fall over.  Why? Buy it out of bankruptcy.  The obvious thing is to re-think the cost base and especially marketing spend.  There is inadequate financial discipline in these companies because capital market spigots have been open – not any longer.  How do you get to break even?  For many of them, the obvious thing is to do what the players in another burgeoning industry of 25years ago – mobile telephony – did: sell to a bigger player who wants the customer base.

7.  BNPL operators are clearly looking to try and be a ‘front-end’ / customer acquisition channel for other financial services providers. Do you see any particular models that make more sense than others?

I’m not a big fan of LFS management but the strategy makes sense – cost share.  I’m not as familiar with guys like Openpay.  The issue of being a front-end for other providers is that today’s world means you can pick and choose individual providers for each function, usually down your phone.  The need for an integrated provider isn’t there.  So in my opinion, it’s about cost synergies not revenue benefits.  Especially in BNPL where the customer base is at the risky end of the spectrum, except for the smarties who leverage the interest-free period of their credit card…..

8. What advice would you give to a beginner investor looking at consumer credit businesses next time around?

Before you look at the specific business, spend time going through a major bank presentation.  Familiarise yourself with how the P&L account REALLY works and the impact of bad debt provisions.  How do these provisions actually work – it’s more against the STOCK of loans rather than the FLOW of credit – but you can pick up the principles.  Understand that bad debts are a cost of doing business in the higher risk consumer credit space, but there are industry metrics and the BNPL players were nowhere near them.  You have to understand the cost base – financial institutions grow costs slowly; BNPL’s went mad.  Don’t commit the same errors the Morgan Stanley Afterpay analysts did, which was to focus on revenue/revenue growth but have no clue as to the cost base and grotesquely underestimate it, and barely talk about it despite elaborate NPV assessments. 

And never forget that very rapidly growing “loan” books don’t do so without some real pain of lending too much to folks who forget to pay it back.

Disclaimer

The correctness of information on this site cannot be guaranteed. This site is for informational purposes only and should not be used for any other purpose, including but not restricted to providing financial advice, decision making or any other purpose. All information is provided "as is" without warranty of any kind, including implied warranty or fitness for a particular purpose.

The information on or produced from this site is in no way to be considered financial advice.

Limitation of Liability

UNDER NO CIRCUMSTANCES SHALL THE OWNER OR ANYONE CONNECTED TO THIS SITE BE LIABLE FOR ANY INDIRECT, SPECIAL, INCIDENTAL, CONSEQUENTIAL, OR EXEMPLARY DAMAGES THAT RESULT FROM THE USE OF, MISUSE OF, INABILITY TO USE, OR RELIANCE UPON THIS WEBSITE OR THE MATERIALS, INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF USE, DATA, BUSINESS OPPORTUNITIES, OR PROFITS. THIS LIMITATION APPLIES WHETHER THE ALLEGED LIABILITY IS BASED ON CONTRACT, TORT, NEGLIGENCE, STRICT LIABILITY, OR ANY OTHER BASIS, EVEN IF THE PERSON OR PERSONS INVOLVED HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. SUCH LIMITATION SHALL APPLY NOTWITHSTANDING ANY FAILURE OF ESSENTIAL PURPOSE OF ANY LIMITED REMEDY AND TO THE FULLEST EXTENT PERMITTED BY LAW.