Description

Buy Diebold 2024 unsecured bonds at ~51 cents to make an equity-like return via a combination of new 2L bonds and 19.99% penny warrants upon closing of the recently announced exchange offer. This trade requires signing up to the Transaction Support Agreement. I think this works because the injection of new money and maturity wall extension de-risks the path to normalized cash flows with recovering supply chains and improved contract economics as tailwinds; non-working capital, fundamental business turnaround is not required. Several exhibits within are from the company’s recently disclosed lender cleansing materials.

I won’t rehash the basics of what Diebold does as there are several other write ups on the company available and much hasn’t changed since then. I will point out that I think the market doesn’t appreciate that Diebold, NCR, and Hyosung essentially form an oligopoly market in ATMs/POS with higher barriers to entry than one might think, particularly on the ATM side. Actors in this space are rational and pricing reflects that. Some may disagree but I do not think there is a wide degree of separation between Diebold products / software vs. NCR. I think the market views NCR as a superior company because they are better managed than Diebold and I don’t disagree with that, but numerous channel checks with bank ATM buyers indicate these two companies are viewed very similarly with purchase decisions driven by geographic availability and existing portfolio mix rather than one doing better on price or having a technological advantage. This is not a winner take all market – Wells Fargo could use NCR machines in Stanford CT and Diebold machines in Boston. As other write ups have mentioned, there is a strong recurring revenue element to ATM installations via service and software components. It’s also worth highlighting that any product or software innovation must go through a six to twelve-month QA period with banking customers before rolling out to market, which provides time for competitors to catch up. Diebold / NCR are emerging from this period of supply chain disruption as net beneficiaries from adjustments to contract economics. Historically, contracts did not provide for annual price increases aside from a service component which had  less-than-robust CPI escalator provisions. New contracts provide 1% to 1.5% annual price increases with pass throughs for labor, freight, and spare parts inflation. I think this is a meaningful development which should allow for structural top-line growth against a flat / modestly declining ATM installed base whereas in the past EBITDA margin expansion was dependent on opex efficiencies. I think uneconomic contracts within Diebold’s backlog will be churned before 1H23.

This week Diebold announced a comprehensive transaction to address its near-term maturity wall and liquidity need by entering into a Transaction Support Agreement with a majority of lenders and bondholders that provides $400mn in new money, refinances the existing RCF with an ABL, exchanges existing term loans into new extended term loans with a partial paydown, and exchange existing unsecured bonds for a new 2L bond / 19.99% penny warrant package. Assuming sufficient participation, closing is targeted by end of December. I think bank debt and secured bonds are attractive trades of their own but will focus on the unsecured bonds which I think are most interesting here.

Through the October 27 early premium date (which I would not be surprised to see extended) unsecured bonds can sign up to the TSA and exchange into new 2L bonds at 95 cents, receive a 5 point early participation fee in the form of additional bonds, and receive penny warrants equal to 19.99% of Diebold’s equity. If you miss the early participation deadline you can still receive warrants, but not the 5 point fee. New 2L bonds will have an October 2026 maturity, pay 8.5% PIK through 7/15/25 with an option to toggle between 8.5% cash / 12.5% PIK thereafter at the company’s option, and are callable at par. The penny warrants are exercisable any time after 4/1/24, will expire five years after the maturity date, and will trade as one CUSIP (bond + warrant package) until the April separation date. The TSA includes a provision that up to $100mn in new equity proceeds may be used to prepay 2L bonds and cancel a pro rata amount of warrants, making the worst case ownership about 15%, but the company is restricted from doing this so long as the new super priority loan and new extended term loans are outstanding. So, I think the risk of having a portion of warrants stripped is low and even if that happens will be an overall credit positive.

Net leverage through the new 2Ls, assuming 95% participation, is 7.9x (7.3x market create) on recently revised 2022 EBITDA guidance and should decline below 6.0x and towards 5.0x as 2023 cash flows begin to normalize (5.3x and 4.7x market create, respectively). I think there are a variety of reasons why $400mn to $450mn EBITDA is achievable here, at which level I think the 2Ls money good at 6x EBITDA, but the fulcrum in a future restructuring. The company projects $470mn 2023 EBITDA and $540mn 2024 EBITDA.

The new super priority loan is not truly super priority – it has a priority first lien on previously unencumbered German collateral and shares a pari first lien with other secured debt on all other non-ABL assets. Additionally, the intercreditor between the secured bonds and term loans will be amended to provide credit enhancement in the form of a second lien on the German collateral and remove any confusion that the collateral packages between these two securities are different, which has been a source of debate. I’m not really sure what the priority lien on the German collateral could be worth as details in the company’s cleansing materials are sparse. However, the new super priority loan is priced at 13% and at market the creation price on the new extended term loans is around 20% so that spread should close, likely in favor of the extended loan. The spread between secured bonds and new extended term loans should also narrow as these securities are now essentially the same. I think this sets up for trades in both term loans and secured bonds.

The exchange offer is conditioned on a 95% participation threshold, and as of release 78.8% of term lenders, 89.7% of secured bonds, and 59.3% of unsecured bonds have signed up. The transaction is quite coercive to disincentive holdouts. Holdout term lenders appear to have substantially all liens stripped and risk becoming structurally subordinate. Holdout unsecured bonds will also be subordinated with the grace period of their coupon extended indefinitely through maturity. The TSA restricts the company’s ability to prepay holdout securities before their maturity date. This risk / reward doesn’t make sense at the term loan level, so I expect most holders to sign on to the deal. Holding out as an unsecured bond could make sense, but I think the economics of the new second lien package are more compelling assuming the business plan works out in addition to having structural seniority over any holdouts. I think realistically there could be 15-25% holdouts bonds, but at these levels, and even with some term lender holdouts, liquidity should work.  

Diebold’s credit stress is entirely the result of supply chain disruption and the company’s inability to maintain stretched vendor payables against a cash on installation working capital cycle. ATM/POS units are shipped overseas, the majority of which originate from German manufacturing facilities destined for the North American market. Port congestion and shipping delays have resulted in product waiting around on ships or waiting for final parts to arrive before entering a long shipping queue. Production days have increased from 60 days to above 100. The global supply chain has been rapidly improving over the last six months, but Diebold has faced a liquidity crunch as they’ve attempted to keep vendor payables stretched – days payable were over 100 as of 2Q22. In fact, 2022 EBITDA guidance was recently taken down by about $40mn to $293mn – this decline was driven by delays in shipping products as certain vendors began demanding cash up front. These products should be shipped 1Q23. New money from the transaction will in part allow the company to normalize their vendors and revert their working capital cycle back towards 60 days in 2023 / 2024.

Additionally, I think the company has provided sufficient information to conclude that absent inflation, supply chain delays, and FX 2022 revenue would have grown by about 1% and EBITDA would have been approximately $400mn.

The company thinks they can do $470mn EBITDA in 2023 and $540mn EBITDA in 2024 with unlevered free cash flow between ~$300 - $400 million. I don’t think many heroic assumptions are required to become comfortable with $400mn - $450mn EBITDA as supply chains normalize / inflation is passed through / price increase take effect. There is a large cost savings component to the company’s figures between headcount reductions and procurement efficiencies. I’m skeptical they can improve procurements, but rationalizing employee overhead should be doable given how bloated the organizational structure has been. FTI has been working with the company over the last several months in developing both their cost savings plan and working capital normalization strategy. EBITDA between 2017 and 2021 averaged about $450mn for context, though I acknowledge a very fair amount of restructuring / other add-backs in historical numbers. An additional tailwind that should drive these numbers is the ramping of DN Series ATMs at the company’s Ohio facility. Most DN Series, their higher margin product suite, is produced out of the German manufacturing facility with North American supply shipped overseas. About a year ago the company began adding additional capacity to the Ohio facility to handle up to 80% of North American DN Series supply. These upgrades are complete, and production should be run-rate beginning 1Q23. Management has not quantified this benefit, and I also think it is somewhat underappreciated by the market, but this move will take out at least 60 days of the product cycle for these units yielding margin and working capital benefits.

As mentioned, I am intrigued by this situation because the path to cash flow normalization seems dependent on rationalization of the supply chain, which we can see is improving, and not taking a view on management’s ability to turnaround non-working capital business fundamentals. This company had already been in a business turnaround between 2019 and 2021, which was working. In my opinion the new money and maturity extension de-risks the path to normalized~$100mn levered free cash flow on $450mn EBITDA, at which point the new penny warrants have substantial value as secured debt should be refinancable via a regular way capital markets process.

Diebold and NCR have historically traded with a levered free cash flow yield between 10 – 20% - these levels imply equity value between $500mn and $1bn and enterprise value between $3.3bn and $2.9bn if levered FCF is ~$100mn by 2024. I think Diebold is probably worth a $3bn EV with a fair multiple range between 6-7x. At these levels, penny warrant value has a PV-25 between approximately 15 and 20 points which seems fair at a 15-20% levered free cash flow yield. A more conservative approach would be to see what the implied warrant value is based on the current share price, about $2.21, which equates to about 10 points of value to 2L notes on a diluted basis. In either case, I think the equity optionality of the warrants has value.

With penny warrant value between 10 – 20 cents, the implied YTM of the 2L bonds credit component is between 40 – 50%, which is too cheap and is a recovery value price. At a range between 15 to 25% the credit component of these bonds is worth 60 to 80 cents and on a package basis 75 to 96 cents assuming 15 cent warrant value. I think it makes sense for there to be a period of price discovery as the market digests this transaction, but right now you can buy 2024 bonds at about 51 cents with limited liquidity. Pricing should rationalize as people do the work, and even more so as additional creditors join the TSA and closing becomes more certain.

I think the argument for holding out 2024 bonds is that if restructuring risk is off the table and these projections are sober the bonds are either money good or will be repaid at maturity, and I don’t think this is an unfair take. There is about 63 points of upside to holding out bonds here between principal appreciation and coupons, the later of which are PIKd through maturity, so about a 2.25x money multiple with a very good IRR. While the structural seniority of the 2L bonds is nice, the reality is that it wouldn’t take a large change in EV to give some recovery to holdout 2024s in a restructuring, assuming 2Ls are money good, which is almost certain to be a large valuation fight. That said, the credit enhancement to the 2L bonds is worth something, and between coupon and warrant value the total money multiple on 2Ls should be around 3.25x through maturity with a ~35% IRR – I think 2024 holders are getting very well compensated for signing up the TSA.

The downside risk here is that somehow the TSA blows up, new money isn’t received, and Diebold files bankruptcy. Bonds will almost surely gap down 20-30 points on that. This could also present an interesting opportunity to buy bonds because we already know the medium-term cash flow outlook, but the shape of new money would probably look more like an equity rights offering, 2024 bonds are probably fully equitized, and secured debt partially equitized with takeback paper. There’s an argument to just filing the company now to realize more material de-leveraging, but I think that’s unlikely to happen given the amount of creditors already signed up for the deal. 

Catalyst

Exchange offer closing

Business momentum