Description

Thesis summary: Toshiba (6502), the flailing and troubled Japanese industrial conglomerate, is highly likely to sell its retail printing subsidiary, Toshiba TEC (6588), half of which is traded publically in Tokyo. Despite the questionable growth prospects for each of TEC’s segments and secular decline concerns, there is a very strong financial argument for acquisition related to a bloated and inappropriate cost structure which any half-decent PE shop (or strategic) should be able to extract, even in Japan. All of this is likely to be concluded in the near-term (<6mos) given the underlying capital issues at the parent and since half of Toshiba TEC remains listed, I think this represents a pretty compelling opportunity to profit (on the long side) from the Toshiba meltdown.

Toshiba TEC has a ~$1.6bn market cap, trades in Tokyo around $5mm/day, and half the shares float, so it is plenty liquid enough for most funds. I should mention that this in no way conflicts with my ongoing bearish thesis on the Toshiba parent company, please see the thread on VIC for recent thoughts there. I think this trade works with or without a short on the parent stock (6502).  

This write-up is divided into three parts: why TEC will get sold; what a financial acquirer could do with the business; and thus what the company will get sold for.

Why TEC will get sold:

It is no surprise that Toshiba is in a boatload of trouble. I will not rehash the whole story here – please see the other VIC thread on Tosh for that – but they basically have a huge capital/equity shortage caused by a) massive losses related to a multi-year accounting scandal and b) more recently, huge cost overruns in their nuclear business. The current equity ratio for the company is a lowly 7%, but this is likely to drop near zero – ie, the company could be theoretically negative equity, a big deal in Japan – due to the losses about to be taken in the nuclear division. Normally a company in this situation – with still a $9bn+ market cap  - would bite the bullet and issue a huge amount of stock, but Toshiba is in the uniquely unfortunate position of being on the Tokyo Stock Exchange ‘watchlist’ (ie, penalty box) and are effectively unable to issue equity into the public markets. This was also the reason why last year – when the capital needs were still significant but not as urgent – the company decided to sell some strategic shareholdings, as well as its medical equipment unit (at the time, one of its few crown jewels).

At this stage (almost 2yrs into the crisis), management is running out of options, having been put 'on blast' by both the stock exchange, its own bankers (who are downgrading internal credit ratings), external ratings agencies (Tosh is now a CCC credit at S&P!) and its own shareholders (who are suing them). Management recently announced a plan to sell 20% (but not more) of the NAND flash division – one of the few ‘good’ businesses left. This plan will likely raise 200-300bn in cash, and perhaps 150+bn in equity capital. I also think the banks will play ball and allow the company to swap around 300-400bn of bank loans into preferred stock (that counts as equity) before fiscal year end (Mar 31).

The issue is the recently rumored scope of additional losses in the nuclear division – 600-700bn – combined with their current paltry equity position mean that even selling down NAND and doing a debt-for-equity swap are not enough. As constituted, they would effectively just cover the additional nuclear losses and still leave them with a mid single digit equity ratio and limited ability to sell stock – along with lesser ownership of the only cash cow business they have left (NAND).

As such, I think it is highly likely that management basically empties the cupboard and sells whatever they can that would raise equity capital. This last point is important: Toshiba is not short of cash (thank you, Japanese banks) but is short of equity, meaning they are highly incentivized to sell anything and everything that would garner a large premium to book value (management has basically hinted as much in recent public statements).

While there are a number of other things likely to be sold (the Elevator business, some real estate, other listed stakes, etc), the rationale to sell Toshiba TEC is particularly strong:

-          It is undeniably non-core (retail POS systems and MFPs/inkjet printers) and the businesses carry low profit margins

-          It is a set of businesses arguably in structural decline

-          It is already a listed entity with discrete management/accounting/corporate, etc so easy to split off

-          it trades at a large premium to stated book equity and so selling would crystallize a sizeable equity gain

This last point is really the key. Since Toshiba already took ~90bn in impairments last year (mostly intangibles + goodwill related to the IBM POS business that TEC acquired in 2012), equity book value is only 63bn, versus current market cap of ~175bn, suggesting large (pre-tax) equity gains if they can sell the asset. This is not the case for basically all other listed stakes – they are either more related to core businesses, trading near book value or only at a smallish premium, and/or are not of meaningful size to create needed equity capital. Simply put, Toshiba TEC checks all the boxes a desperate Toshiba management is looking for at the moment.

How much can they sell TEC for?

This brings us to the potential value for TEC. Toshiba TEC has two main businesses: POS systems, receipt printers, self-checkout systems, etc (basically all the hardware you see at a CVS/supermarket/retail outlet), which is about 60% of sales; and retail/office printers (mostly MFPs and some inkjet printers), around 40% of sales (note - the company's segment reporting is NOT consistent with the product breakdown, so analyzing each segment separately is less helpful than on a product-mix basis).

I am not going to make an argument that these are ‘good’ businesses because they are debatably in structural decline and TEC margins are much lower than comps. However – the retail POS business – mostly IBM’s acquired business – still has somewhere between 25-30% of global share and is the market leader; furthermore capex needs in the retail hardware side of the business should be very low (competitors NCR and Diebold spend 1-2% capex/sales and these companies have banking/ATM exposure which is more capex heavy). Similarly, these companies spend only 3-4% capex/sales, additionally suggestive of end markets in secular decline and not needing excessive new investment. Despite this, Toshiba TEC has consistently spent 6% of sales on R&D and 2-3% of sales on capex. That is to say, TEC has consistently outspent their main comps (in the POS space) by 300-400bps over the last five years (at least) – and there is no massive scale difference either (NCR sales are $5-6bn, TEC is ~400-500bn JPY, and Diebold is ~$2-3bn). It is hard to deny TEC appears bloated and inefficient in their investment spend and thus, a financial (or strategic) acquirer should look at this and immediately expect to extract significant synergies.

 To be fair, 40% of TEC’s business is in printers, where capex and R&D needs run higher (looking at comps such as Epson, Konica Minolta, Brother, etc). But looking at TEC profitability on a total opex/sales basis (ie, the difference b/w gross and operating margins) versus a variety of POS and printer comps suggests still a very inefficient cost structure that cannot be explained entirely by scale, in my view. On this basis, TEC is running ~10pts higher opex structure versus peers (a composite average of Konica Minolta, Brother, Seiko Epson, NCR, and Diebold) explainable partially by higher R&D and inefficient operations and perhaps some scale benefits (at the higher end) but not by gross margins (where TEC is ~15pts north of POS peers and only ~3pts on average south of printer peers).

Another way to consider TEC’s underperforming profitability is to look at the discrete A3 copier market. In this segment, TEC still has small but meaningful global share (around 7% on a volume basis and 5% on a value basis) and I think A3 copier sales represent the bulk (~60-70%) of their entire printer sales, hence this should be somewhat representative. The difference between volume and value share of course implies TEC sells at a discount to larger, better offerings – which makes sense – but consider Sharp: it too sells at an implied discount (Sharp has 9% volume share, 7% value share in A3 copiers) but reported Sharp OP margins in their copier business have been in the high single digit range for years. Compare this versus TEC Printing Solution margins (which should include some higher OPM retail peripherals) which have averaged 4-5% and only reached 6% last fiscal year. Clearly given market shares, the absolute scale of these two businesses is similar, so again, I think the takeaway is there is considerable scope for cost cutting if/when this business is acquired.

Assessing current + acquirer’s EBITDA:

I think there will be a reasonably quick bounce-back to TEC earnings this year (to Mar'17): last year included a number of one-time items (other than the intangible impairments) which should improve the cost profile somewhat (eg, inventory writedowns). Additionally TEC closed 8 sales facilities (out of a total of 36 in FY15) throughout Europe, and fired 100-200 employees (still small versus 20k employee footprint). There is still clearly a lot more to do, but I think OPM can get back to 2.7% for the full year (1H OPM recovered to 2.6%) which would basically imply EBITDA around 29bn JPY for FY17E – this would put the stock at current mkt (174bn mkt cap, 143bn EV pre pension) at ~5x EBITDA – not expensive in absolute terms and the low end of Japanese comps, but admittedly not exciting for this business mix.

From here we can assess how much an acquirer could extract to increase EBITDA. Here I am basically assuming a financial buyer given a) Tosh is looking for a quick sale; and b) there could be anti-trust concerns in selling the Retail POS business to another large player. Having said that- I think a number of Japanese strategics could at least look at the Japanese businesses and there would be additional synergies if that happened.

My assumptions are the following (bear to bull cases, respectively, from the acquirer’s perspective):

-          No top line synergies

-          100-300bps cut to R&D spend

-          5-10% reduction in personnel costs

-          0.5-1% additional opex cuts (corporate, HQ reductions, end public co costs, etc)

-          An increase in capex to match long-term D&A rate

-          60/40 debt/equity funding split and 5-2.5% cost of debt

-          40-20% equity premium versus current

I do not think these are aggressive assumptions, especially given normal PE cost reduction methodologies, but I assume some kind of ‘soft’ restructuring given it is a Japanese company. The capex number, as well, could probably come down as D&A should run higher than capex given the profile of these businesses. Nevertheless, even in these cases, the implied purchase multiple in the base/bull cases (ie, 24-33% upside from here) to a purely financial acquirer is very attractive at around 3-4x acquirer’s EBITDA:

Post-restructuring implied EBITDA margins around 10-11% also do not look aggressive when compared to extant Japanese peer printer margins (in the 10-14% range). The scale of Yr1 cash returns, 22-34%, versus implied leverage (around 2-2.5x) is also quite low for this kind of LBO/MBO transaction (admittedly this is not a ‘proper’ LBO model approach where you consider out-year returns and exit multiples, but I just want to demonstrate the absolute cheapness even with quite conservative day-one assumptions).

KKR has been active in Japan, recently buying Calsonic Kansei as well as Hitachi Koki – a quick look suggests they did so for much lower extant cash returns and perhaps more difficult restructuring (though I haven’t looked in detail to be fair). Buying TEC would not be a huge transaction (~$2-2.3bn USD) and is eminently doable for KKR or any other large buyout shop (or even clubbed with management/local partners). Others have commented (eg on the SNOW thread) re the availability of PE capital, the shortage of attractive deals, and the availability of credit markets to fund these kinds of transactions. These factors too speak to the likelihood of a financial buyer stepping up to the plate here.

Risks:

There is risk that Tosh sells their 50% stake and the entity remains listed – however given the need to conduct further cost-cutting and likely headcount restructuring, I would think the preferred method would be to do so out of the public eye. Normally, when the major shareholder sells their stake, minorities are offered a tender at the same price (this is what happened in the recent Calsonic Kansei and Hitachi Koki cases).

In the event TEC does not get sold, I still expect earnings to post a recovery, admittedly from a low base, such that you are buying a relatively bad collection of businesses at a low multiple with some likely near-term improvement in the story. Of course this would not be ideal but I don’t think it would be a disaster. And I do really think this gets sold.

In sum it would not be a stretch to see this ultimate sold for another 25-40% upside, and I think this all happens very near-term (perhaps 3-6mos).

Catalyst

Nikkei reports TEC for sale

Business gets sold