BFF Banking Group: Expect Delays

Based in Milan, Italy, BFF Banking Group (Banca Farmafactoring, “BFF”) provides non-recourse factoring and lending services to the suppliers of public healthcare authorities, public administration entities, and small and medium-sized enterprises. It was founded in 1985 with an initial investment of EUR 260,000 by a consortium of pharmaceutical companies to better manage their cashflow amidst the Italian public sector’s notoriously long payment delays. The bank purchases trade receivables at a discount and then manages their collection. This arrangement allows BFF's clients to receive immediate cash, while it makes money on the discount, late payment interest and other fees.

BFF is far from the only lender in Italy or Europe to offer factoring services. Still, its focus on the public sector - particularly healthcare - has allowed it to build a lucrative niche. From 2003 to 2022 - a period which included the Global Financial Crisis, the European Debt Crisis and COVID - it grew its receivables book by 10% p.a., delivering an average return on assets and equity of 2% and 24%, respectively. And from the end of 2002 to the first quarter of 2023, it increased its net worth from EUR 56 million to EUR 800 million while paying a total of EUR 1,064 million in dividends.

EUROPE’S Factoring MARKET:

Factoring is a valuable commercial service. It allows the vendor of the receivables to collect cash early and redeploy it for better use elsewhere. And compared to a working capital loan, factoring relieves the vendor of the administrative burden of collecting late payments. Specialists like BFF are far more familiar with the nuances of the public bureaucracy and legal system, and how to find solutions and settlements. They are also sensitive to the vendor’s ongoing relationship with their clients, and know how best to manage a difficult process amicably.

The EU Federation of Factoring & Commercial Finance, a trade body, estimated turnover in Europe’s factoring and commercial finance industry in 2021 totalled EUR 2 trillion, representing 11% of GDP. The largest markets that year were France (EUR 365 billion turnover in 2021), the United Kingdom (EUR 328 billion), Germany (EUR 309 billion), Italy (EUR 258 billion) and Spain (EUR 199 billion).

Volumes - the notional value of receivables purchased in a given period - grew an estimated 7% p.a. from 2010-2021, driven by the 13% p.a. growth of non-recourse turnover, which now comprises 53% of the market (note: non-recourse means that the purchaser of the receivables takes the full risk on their collection). The average volume per client stayed about constant during this time at EUR 7.7 million, but the number of active client relationships doubled from 126k to 265k and the average days outstanding increased from 43 to 64.

BFF’s PUBLIC SECTOR NICHE

BFF was created to serve a very specific niche within this larger market: the suppliers of the public sector.

Late payment is a particular issue for the Italian public sector because regional and municipal governments are responsible for providing public services but have limited powers to tax and raise revenue. They are, therefore, dependent on the central government for fiscal support. But the central government’s own financial difficulties make this support irregular and hard to come by.

And so it reverberates down the chain to businesses: data from Assifact, the Italian association of factorers, shows that in March 2023, receivables owed by local governments and healthcare authorities comprised 54% of total receivables owed by the public sector but 69% of overdue receivables. Cashflow is a particular concern to the suppliers of these bodies, making factoring a particularly attractive alternative.

Assifact has 31 Associate members who comprise “the substantial entirety” of the Italian factoring market. While most of these firms are much bigger than BFF, BFF dominates its niche. If we make a crude assumption that all its Italian volumes went to the public sector in 2022, then BFF accounted for 23% of public sector receivables. The company has much more granular data on its share of specific verticals, as shown below in this slide from 2Q18, where it states it had a 35% share of public healthcare receivables and a 33% share of local government receivables.

EU Late Payment Directive

Italy is not alone: chronic late payment by the public sector is such an issue in Europe that the European Union intervened to protect businesses by legislating penalty interest on overdue receivables. This penalty interest was set in 2000 at the European Central Bank refinancing rate + 7.00% p.a. and raised in 2011 to the European Central Bank re-finance rate + 8.00% p.a. It applies to all member states of the European Union and begins to accrue on public administration receivables older than thirty days, and national health service receivables older than sixty days.

We can see from BFF’s own data that after this penalty was introduced in 2011, it had a profound impact in Italy on days sales outstanding (DSOs), reducing them from 350 to below 150.

Nonetheless, while repayment of the notional principle has improved, collecting the late payment interest still takes a long time. Perhaps because no interest is charged on the interest, public authorities do their best to delay payment for as long as possible.

For accounting purposes, BFF’s most recent assumption is that it will take 2,100 days on average to collect half of the late payment interest it is owed - that’s five and three-quarters years! The reason it only expects to collect half of what it is owed during that time is that it must often settle with debtors to get paid at all.

BFF’s ECONOMICS and advantages

This is a good point at which to discuss the economics of a typical receivable.

There are two components to the transaction: a) the discount on the notional amount; and b) the expected amount of late payment interest, which is a function of i) how long it takes to collect the notional, ii) the current ECB refinance rate and iii) the estimated collection rate. I have used BFF’s assumptions for collection and the minimum EU rate for late payment interest to show that in this scenario, the receivable earns an IRR of 5.2%.

This is a very attractive return given the risk and alternatives in Europe. The real magic, however, is not in the numbers but in the qualitative insights the exercise reveals:

The EU’s late payment interest rate offers an asymmetric upside because it cannot go below 8%. It is fixed semi-annually in January and July and is likely to be set at 11.75% later this year.
In practice, it takes on average nine months to collect the notional. BFF is therefore regularly repricing its loan book to market rates by way of a wider or narrower discount. Together with the variable late payment interest, this means that BFF takes much less interest rate risk than banks which lend into longer tenor products like mortgages.
Once BFF collects the notional, it can immediately redeploy the capital into purchasing new receivables. It is therefore no longer using its balance sheet while waiting to collect the late payment interest, arguably making the true IRR on that income stream infinite.
Collecting late payment interest is difficult and takes a long time. But the ability to do so transforms this asset from a 2.0% IRR to a 5.2% IRR (per my assumptions above). Put another way, the ability to collect is a barrier to entry: to compete with BFF’s pricing, competitors need to be confident they will survive long enough and collect as much as quickly.

A sidebar on accounting: BFF reports late payment interest and recovery costs on an accrual basis using the above estimates (50% collection and a 2,100-day wait) and an assumed 5% IRR. BFF amortises the income it expects to collect over the life of the expected collection period, during which time it sits on the balance sheet as an asset until cash is received and the entry is reversed. Because most accrual income is reported in the first year, profits overstate cashflow while the business is growing.

BFF’s assumptions on collection have proven conservative: in the six years since listing, it has recovered more cash in each year than originally assumed - not a lot more (EUR 81 million), but enough to give comfort on the accounting. As of 1Q23, there is EUR 548 million of potential income which sits off-balance sheet, which is what would be theoretically available to BFF if it could collect 100% of what it is owed rather than 50%.

A second sidebar on expenses: the cost of collection is borne by BFF’s creditors, making the exercise cash neutral, albeit not time neutral.

And a third sidebar on fees: the European Court of Justice ruled in 2022 that public sector creditors owe a EUR 40 processing fee to debtors like BFF for each receivable outstanding. This adds a third income stream to the two mentioned above.

not all leverage is equal

In addition to its attractive unit economics, BFF enjoys operating and financial leverage.

Like an asset manager, there is operating leverage inherent in the model: the cost to recover EUR 10 million is the same as EUR 1 million, but the profits are exponentially greater. BFF disclosed that it processed 1.2 million invoices in the twelve months to 3Q22. If we use that figure as a rough estimate for the invoices issued in calendar 2022, then the average notional was EUR 7.6 million. I would like to know how that’s changed over time, but the company does not release this data.

Then there is financial leverage, which gets a little technical. Banks are required by their regulators to meet certain minimum capital adequacy ratios which govern their leverage and exposure to credit risk. Different assets are given different ‘risk weights’ to account for their likelihood of default. For example, a bank’s holding in a government bond will typically have a lower risk weighting than a loan to a private business. With the same amount of capital, the bank can therefore invest more in bonds than it can in loans. The trade-off is that ‘less risky’ assets typically earn lower yields.

But BFF gets to have its cake and eat it too: because its counterparty is the public sector, its high-yield receivables are deemed to have a low risk weight of just 20% of their notional value. This permits BFF to supercharge its returns by running a high level of straight leverage (it ended 1Q23 with a 15x ratio of assets to equity) while still reporting a very high capital adequacy ratio (it ended 1Q23 with a Core Equity Tier One ratio of 17.0% versus the minimum required of 12.5%).

Furthermore, because the risk of default on receivables owed by public authorities is virtually zero, BFF does not have to provision for credit risk the same way ordinary banks do. This improves its earnings and frees capital for productive investment and dividends (provisions in Italy otherwise carry a punitive risk weight of 150%).

You would be correct to point out that the whole concept of risk-weighted assets is flawed, and I would have to agree with you. Their definition is arbitrary, and there is not enough nuance to distinguish good credit from bad within a category. Moreover, leverage and credit risks are just two of the plethora of catastrophic risks that banks face on an everyday basis. As if to prove the point, Silicon Valley Bank reported a healthy Core Equity Tier One ratio of 15.2% literally weeks before going bankrupt.

This is why it’s so important to understand the nature of BFF’s assets: as I explained above, their short duration means that BFF is not exposed to the same degree of interest rate risk that blew up SVB (which had invested in long-duration government bonds at extremely low yields / high prices). Instead, in response to tightening liquidity conditions, BFF can quickly dial back its loan book by slowing the pace at which it re-invests in new receivables. Or, it can sell other, less attractive assets, which is exactly what it did in 1Q23.

TRANSFORMING THE BALANCE SHEET

BFF obtained banking licences in Italy (2013), Spain (2015) and Germany (2016), permitting it to raise deposit funding, which it does via online channels and third-party brokers.

Then, in 2020, BFF acquired DEPObank in a transformational deal. DEPObank was spun out of Italian financial conglomerate Nexi in 2018, when Equinova, its private equity owner, separated the regulated securities services and payment settlement services businesses from what would become Nexi’s core payments business.

For EUR 239 million in cash and approximately EUR 63 million in equity (14 million shares), DEPObank added EUR 86 million in adjusted net profit before tax to BFF in 2022 (42% of its total), all of which was asset-light fee income.

Crucially, DEPObank also gave BFF access to cheap funding via the EUR 7.5 billion float it had otherwise deposited at the ECB or invested in sovereign bonds (the benefit of these lower funding costs are reported under BFF’s Corporate Center business unit).

After the acquisition, BFF’s receivables book fell from 67% of assets in 4Q20 to just 25% in 1Q21. I would be very happy if BFF could invest more in receivables on the terms described above, but it is unfortunately constrained by the willingness of businesses to sell. The receivables book had grown to 43% of assets in 1Q23 and hopefully reverts to its earlier level in the coming years.

DEPObank’s float also allowed BFF to retire bonds and wind down deposits. However, the float has shrunk recently as clients choose to invest idle deposits in higher-yielding money market funds. The securities services business also lost a major client in 3Q22.

BFF has had to revert again to more expensive deposit and wholesale funding to fill the gap, which led to a 7x increase in its interest expense in 4Q22 and a 6x increase in 1Q23. This is an important point: while BFF’s assets reprice quickly, its liabilities reprice even quicker.

CYCLICALITY and INTERNATIONAL Expansion

Demand for factoring follows the credit cycle. When there is abundant liquidity in the market, public authorities pay their bills quicker. When liquidity is tight and macroeconomic uncertainty heightens, clients become more eager to sell their receivables and collect cash immediately.

BFF CEO Belingheri put it this way in an interview in 2018 about macroeconomic risk in Italy: “Fragile markets are good for us: abundant liquidity over the past few years has made our business less attractive, while now companies need to assess if it makes sense to hold credits with a potentially longer reimbursement time. The higher risk perception is good.”

Looking at Assifact data for factoring volumes, demand - perhaps counter-intuitively - declined during 2020 and 2021 when Italian governments were flush with COVID-stimulus.

Fortunately, BFF has been able to expand internationally over the last two decades to diversify and grow against the cycle.

It responded to its clients’ needs by expanding into Spain (2011) and Portugal (2014), and later sought new opportunities in Greece (2017), Croatia (2018) and France (2019). It acquired healthcare lender Magellan in 2016 for EUR 103 million, giving it a footprint in Poland, the Czech Republic and Slovakia.

BFF is in the early days of building its franchise outside of Italy. As explained in the slide below, it takes time to earn the trust of potential vendors, who must trust that BFF will properly manage the relationship with their clients and not lose future business.

However, the public sector in these other markets is just as challenging to deal with as Italy, and they should deliver well over time. BFF reports the days sales outstanding (“DSO”) of its receivable book. In 2022, from lowest to highest, they are Croatia (52), France (79), Italy (132), Spain (132), Portugal (175), Greece (248) and Slovakia (785).

ALIGNMENT AND INCENTIVES:

The pharmaceutical companies that founded BFF sold to private equity firm Apax Partners in 2007, which in turn sold to private equity firm Centerbridge Partners in 2015. Centerbridge took BFF public in 2017 through a partial sale of its stake (i.e. no new capital was raised), selling further tranches in the market until it had exited completely in 2021. Equinova, the former owner of DEPObank, finalised the sale of its stake in BFF in March 2022.

This leaves BFF’s CEO Massimiliano Belingheri as the bank’s single largest shareholder, holding just under five per cent of shares outstanding. Belingheri joined Apax Partners in 2001 after graduating from the Harvard Business School. He became a partner in 2008 before leaving in 2013 to become BFF’s CEO.

I read twenty years of BFF’s annual reports and liked what I saw. But to put it bluntly, I would not have invested if its CEO was not its single largest shareholder, and if I had not seen him making substantial purchases of stock using his own money, as he did in February 2021 when he bought 3.5 million shares in an off-market tender at a price of EUR 5.45/share.

If there’s one thing to remember about 2023, it’s that banking is a perilous business: mixing credit, duration and liquidity risks with a leveraged balance sheet can see shareholders’ equity vaporise overnight. Managers’ short-term interests can easily diverge from long-term shareholders’: today’s reported earnings all too often become tomorrow’s non-performing assets and mark-to-market losses. So, as minority shareholders and outsiders, all we have to rely on are management’s track record and their word.

Belingheri was Italy’s highest-paid banker in 2022 thanks to stock options awarded earlier in his tenure. We can debate the fairness of that pay, but I would suggest that he has created a tremendous amount of value since he took the job in 2013.

Because Belingheri is a minority shareholder himself, his compensation package can only be approved with other shareholders’ support. So, it’s worth noting that his proposed package was rejected by shareholders in 2022 before being amended and approved in 2023. Personally, I would have preferred to see all the financial metrics quoted on a per share basis. Incorporating a cost of capital in the calculation of his long-term incentives is a step in the right direction but is too narrow to capture all the risks a bank faces.

VALUATION:

BFF listed in 2017, which doesn’t give us a lot of trading history. It is currently priced at 2.2x trailing price to tangible book, which is slightly below the median of its range.

One thing to bear in mind is that Italy has a 26% withholding tax on dividends. I have asked BFF’s IR manager why they don’t return capital through buybacks instead and apparently the banking regulator’s rules are simply too complex.

WRAPPING IT UP:

A lot of people will say that it’s difficult to do business in Europe. Or that Italy’s parlous public finances and unstable governments make it too risky to invest there. In a way, BFF’s success proves them right. I struggle to think of many balance sheet-led financial services firms which can boast the longevity and level of BFF’s high returns on capital.

The company’s almost forty-year track record demonstrates two things. First, that BFF is playing the right game. Businesses which supply the public sector in BFF’s markets wait a long time on average to get paid, making factoring a valuable alternative to manage their cashflow. Second, BFF’s management has an exceptional ability to manage risk and make money. They’ve grown while returning an extraordinary amount of capital. And they have made some savvy deals too. BFF’s competitors have not performed even remotely as well.

Of course, like most banks, BFF’s recent earnings show the painful impact of the rapid tightening of liquidity. The cost of money has gone up, and it has yet to fully reprice its assets. There is latent earnings power here that should become apparent in the next twelve months.

I have not covered BFF’s service business in depth here because they are secondary to the factoring business. I may come back to them in the future.

Disclosure: At the time of posting, clients of Longriver Investment Partners Limited owned shares of BFF. This may have changed at any time since.