You have to be prepared to look in some strange places to find cheap stocks these days, and one of them is small-cap Japan. This can be a pain for a number of reasons (language, filings/disclosures, liquidity, etc) but what’s a guy to do? The S&P chart looks like this:

Meanwhile the Topix 1000 (broad-based Japanese stock benchmark) looks like this:

This is only a 1yr chart, but I could pull them both back over 5yrs (or 10yrs, or longer) and show you that Japan has massively underperformed the US rally. There are many good reasons for this – lack of buybacks, lack of aligned managers, structurally lower RoEs, structurally lower growth, etc etc – and this will not be an examination of those issues. Rather, this is simply to say that there are a) lots of cheap stocks in Japan (so a good place to look for bargains in an expensive investing world); and b) Shinoken (8909.JT) is one of the cheapest and most attractive of the bunch, in my view.

So this is really just a post about one really cheap stock, Shinoken, that happens to be a Japanese small-cap. If that is not for you…no problem 🙂

Meet Shinoken

Shinoken is a Tokyo-based real estate developer. They focus mostly on apartments (in the Japanese context this means low-rise buildings of 2-3 floors), though they also do condominiums (high-rise, higher-ticket projects, called ‘mansions’ in Japan). They offer a fully turnkey solution where they select the land plot, buy the land; design the building, etc; construct the building (often using their own in-house general contractor); deliver the building to its owner, post construction; and then offer a full suite of management services for the owner to make their life easier (property management; rent guarantee; insurance; electricity/gas bill management; etc). The English-language IR section of the Shinoken homepage has some decent information on the company and its operations (of course the Japanese site has much more information).

Its worth spending at least a moment on the individual reporting segments at Shinoken because a key tenet of the investment thesis is that an increasingly large % of earnings comes not from pure development, but from ancillary services that are more recurring in nature (and thus would expect to garner a higher multiple from the market).

Business segments:

Development (61% of LTM revenues, 60% of LTM segment EBIT): development and sale of apartments and condominiums in select cities, as described above. Apartment and condominium have different business characteristics and capital requirements, but are lumped together into one segment (doesn’t make it easier for us, does it!). As you might expect, this is a lumpy number (both top and bottom lines) and has been under pressure in recent quarters due to market-wide pressures (more later), but has grown strongly over the longer-term.

Management (14% of LTM revs, but 25% of LTM EBIT): by nature higher margin recurring revenue, principally driven by the % of rents taken by the management company (around 5% of rents usually) in return for managing all aspects of the rental process on behalf of the landlord. EBIT in this segment has grown consistently in recent years whilst scaling (now printing low 20s % EBIT margins), as the company has consistently added services to its offering (rent guarantee, insurance, etc) and signed up more landlords for coverage as it delivers new units from its development pipeline. This business is extremely asset-light (simply managing more buildings on the same platform) and has highly desirable economic characteristics – as long as they can maintain occupancy in their existing managed fleet (and occupancy is 98% right now), earnings should be relatively stable even if the development business were to stop completely. Assuming development growth remains stagnant, Management EBIT as a share of total company earnings will continue to rise (suggesting improving business quality). My FY19E numbers suggest close to 4bn in EBIT from this segment alone – which even at a low multiple for this kind of quality business (say 12x) justifies close to double the current market cap (27bn today).

General contractor (21% of LTM revs, 12% of LTM EBIT): a small contractor that was acquired a few years ago at a very low multiple of today’s EBIT (around 1.5x run-rate EBIT today) and has expanded significantly beyond simply building Shinoken’s units (though this was the main purpose in acquiring it, to achieve integration and cost efficiencies). Today, ~80% of revenues are derived from external business, and while of course this is at risk related to the overall construction industry, further makes this revenue stream more independent from Shinoken’s core business and thus – from a valuation perspective – should help defray concentration risk in the earnings pie for the company, over time.

Energy and Life Care (combined, 3% of LTM revs and 5% of LTM EBIT): two small segments that I have combined – Energy relates to the provision of electricity/LPG services for Shinoken’s buildings (on behalf of landlords) and is ‘bolt-on’ service business related to the management segment; while Life Care provides assisted living services at a number of facilities throughout Japan (currently almost fully occupied). Many of the tenants are ex-Shinoken residents who have aged out of their apartments – thus providing a further leg to the Shinoken service flywheel. These businesses are both small but again, have consistently grown, and are, once again, somewhat independent of the core apartment development business (ie they would exist as semi-annuities even if development of new buildings stopped completely).

Putting it all together, the important takeaway is that the % of segment EBIT coming from OUTSIDE the development business has been rising, in absolute and relative terms, consistently over the last few years. Here is how it looks over the last 12 quarters (using my estimates for the rest of 2019):

That is to say, non-development EBIT – much more recurring and repeatable in nature – will, by the end of this year, be closer to 50% of segment EBIT on a LTM basis (versus 20% a couple of years ago). Yes, a large contributor of this is the decline in earnings in the apartment segment – which we will presently discuss – but it is also due to the inexorable march higher of the recurring earnings in the service-type businesses (the orange bars above).

To conclude: Shinoken is becoming much more than just a typical real estate developer – its just that the slowdown in the apartment business has masked this (for now). I expect that to change in short order.

Shinoken’s development model

Of course, apartment development in Japan is highly competitive, so Shinoken’s success – and as we shall see they have been incredibly successful – has been built upon a few key advantages:

– niche target customer: Shinoken specifically targets individual salarymen (and women) looking to acquire an investment property on a no/low money down, high LTV basis, for income.

– land/site acquisition: Shinoken targets only very specific sites for development (<10 mins walk to nearest train station) in select cities (Tokyo, Osaka, Fukuoka, Nagoya, Sendai) where the % of single inhabitants is structurally growing. Since the majority of their apartments are smaller units that cater to solo inner-city dwellers, they are therefore running with the demographic tailwind only in certain cities.

– scale, cost benefits: while Shinoken only supplies ~1% of new housing starts in Japan, they are one of the largest providers in their specific segment (apartment sales for investment purposes) given the fragmented nature of the market. Since acquiring a General Contractor in 2014, and growing rapidly thereafter, they have been able to achieve economies of scale in development (eg unified building designs from a set of templates) and construction (vertical integration) such that they can undercut slightly on offered rents (generally they under-price the extant market by 5%) to drive full occupancy and thus increase the perception – for both tenants and landlords – that Shinoken properties are desirable because they are always full.

– financing: unlike some other outfits, Shinoken self-finances development and construction of their projects. This clearly increases asset intensity and limits growth rates given balance sheet constraints, but also allows much faster inventory turnaround times at scale (since they have developed construction capabilities over many years, a new apartment project can be fully constructed in 6-8mos). Most importantly, it also distinguishes the company from other operators who rely on their customers to finance development risk – a reliance that has led to bad behaviour by competitors (as we shall see).

– high occupancy rates: Shinoken has maintained very high occupancy for its buildings (95-98% in recent years, versus the national average in the low 80s %). This is partially due to astute management (they manage the majority of properties they have developed); partially due to pricing (as described above); and partially due to general tightness in the market.

– sales approach/customer acquisition: Shinoken does not cold call (common in the industry); most customers are either repeats or referrals, or garnered from TV advertising.

It is perhaps easier to let the financials do the talking. Over the last 9 years, Shinoken has compounded book equity at a 42% CAGR, growing book value from <2bn JPY (FY11) to 41bn JPY (FY19E) – ie a 20x increase – whilst generating average pre-tax ROICs of 20%. You probably don’t need to research many other Japanese companies (or Japanese real estate companies) to figure out that is exemplary by any measure:

Putting aside questions of sustainability, normally you would expect to pay a substantial premium to book value for a company compounding at 40%, right? Conceptually, if we thought even 30% returns were sustainable, the implied P/B metric (assuming say a 10% cost of equity and no growth) would be 2x. How has the market valued Shinoken over the years?

Clearly Shinoken isn’t getting much credit. The stock trades at 0.7x book value today – well below historic averages (1.9x over this period) – and effectively implying a well-below cost of equity sustainable return going forward (maybe in the 6-7% range, judging from the 0.7x price/book multiple today). This is despite the company having printed 40% compound for almost a decade, and still printing 25% last year and 20% this year (in the face of significant delevering of the balance sheet)! In absolute terms, this seems somewhat crazy, and it is only in reference to the poorer returns (but still extremely positive and well north of cost of equity) of the last couple of years that we can begin to understand what the market is thinking.

Simply put, the market is telling you at this price that Shinoken’s model is broken, and that it will revert to a below-average earner in the very near future. Before digging in to why I think that is completely wrong, it’s worth examining what has changed in the apartment development industry in the last year or so.

 

An annus horribilis for Japan’s investment property developers

The last 18 months have been incredibly rough for the Japanese investment property industry, as the sector has been beset by a number of damaging scandals. In no particular order of importance, here are some of the main issues that have encountered the industry since the beginning of 2018:

• the Suruga loan scandal: Suruga Bank, a regional bank based in Shizuoka prefecture, admitted to a wide-reaching scandal involving the forgery of loan applications for ‘share-house’ schemes (shared home ownership and co-habitation, often marketed towards single women). Simply put, bank executives had faked income and other data to support loans to unworthy borrowers, all to support unrealistic lending volume targets. Starting from April last year, the scandal culminated in the resignation of five senior bank executives (including the Chairman and President) in September; and Suruga ultimately being forced to seek assistance from Shinsei Bank, in May of this year after admitting that losses on illegitimate loans could top 1 trillion JPY. This FT article presents a reasonable summary of the issue – the main takeaway is that this has been an overhanging issue on the sector for most of the past 12+ months;
• the Leopalace construction scandal: beginning in May/June last year, Leopalace, the builder and manager of over 30,000 apartment units in Japan, admitted that they had not maintained necessary building standards in the construction of some 200+ apartments (violating fireproofing and soundproofing standards). Shoddy construction techniques were likely employed to try to save costs. By February of this year, the number of shoddy units had been upsized to ~1300, and Leopalace was paying to move 14k residents out of these units to conduct repairs. Needless to say, Leopalace’s new construction business has collapsed; their occupancy rates have fallen; the share price has cratered; and many are wondering if the company ultimately survives (they have a levered balance sheet now given the remedial costs and, newly, a terrible reputation). Clearly this has been a further ongoing impediment to sentiment in the sector;
• the Tateru loan scandal: perhaps most damaging for Shinoken directly, however, has been the Tateru scandal. Another property developer, Tateru admitted to falsifying loan documents on behalf of its customers (inflating incomes, forging bank balances etc), to achieve loan approvals from the banks. The scandal broke in September last year, and grew out of the Suruga scandal (in that a brighter light was being cast upon the industry, along with a surge in third-party investigative journalism). By June of this year – and despite the fact that Tateru’s business has fallen to near-zero as a result of the scandal – the government slapped a ‘Business Improvement Order’ against the company, effectively making it very difficult to continue operating in this particular line of business. Again, this was a long drawn-out affair, only really resolving itself by June of this year (hence, a further multi-quarter overhang on sentiment).

 

The cumulative impact of these three large scandals – involving banks, property developers, and constructors – was clearly to cast a pall over the entire sector. Tateru – one of Shinoken’s closest competitors – saw its stock price go down 90%, of course dragging down Shinoken by association. Meanwhile, all the banks (not just Suruga) have given the cold shoulder to apartment lending, decreasing loan volumes and tightening lending standards aggressively to make sure they don’t get caught up in the maelstrom.

Given the extent of the scandal, it has been hard for the market to assume (in the short-term) that other players in the space are operating differently, and this is why a company like Shinoken is available today at <4x earnings. But let’s take a closer look at the actual content of the issues, and why Shinoken may in fact have been the proverbial baby thrown out with the bathwater.

1) The Suruga scandal – the core of the issue here was inappropriate lending against shared-houses (co-habitation apartments). Shinoken does not, and has not, constructed these kinds of apartments. Moreover, Suruga’s home market (Shizuoka) is an area Shinoken does not have meaningful (or any) business, and furthermore, Shinoken confirmed subsequent to the Suruga scandal that they have not had more than a single mortgage issued by Suruga in the last three years, and that Suruga is not one of their key lending relationships. Conclusion: in the absence of any direct linkages, it is hard to see how Shinoken could be directly entangled in this particular imbroglio.

2) The Leopalace scandal – the issue here was shoddy construction and breaking building code standards in Japan. A harsh light has been shining on the industry for the better part of 1-1.5 years now, after the Leopalace issue first broke, and neither Shinoken nor its general contractor have had any issues crop up at all regarding quality. In fact, they appear to have maintained a very strong reputation for safety and trust with their tenants, as evidenced by the ongoing high occupancy rate (98%) of its buildings. We know that Leopalace’s shoddy construction was related to an inability to constrain costs – but Shinoken strategically brought a contractor in-house to achieve cost synergies through integration. Furthermore Shinoken does not try to maximize construction profits (gross margins in the apartment segment are lower than comps) and instead tries to focus on turning inventories faster and increasing the unit fleet to expand the base for recurring services). Conclusion: while in theory this could always crop up at some point (since I haven’t personally inspected the construction quality of all their buildings), if it hasn’t been an issue thus far with so many eyeballs focused on it, it seems unlikely to be a trigger event now. Furthermore, there are business model differences between Shinoken and Leopalace that could somewhat explain why these issues would be less likely to occur at Shinoken.

3) The Tateru scandal: Tateru is most similar to Shinoken in terms of business area (targeting low or no-money-down apartment sales for investment purposes), but utilized a different model to achieve higher growth unconstrained by its own balance sheet, and this is where the issue of moral hazard occurred. They relied on their customers to finance land acquisition, development and construction of the apartments (as opposed to Shinoken, which financed all development from its own balance sheet) – and they ‘helped’ their customers fill out all the necessary documents to get loans to allow purchases to occur. Clearly, in pursuit of growth, this model introduced temptations to Tateru employees which they could not resist. With regard to Shinoken, the company has maintained – post the Tateru scandal breaking – that they have far fewer opportunities to even handle customer loan application data; that they are not really involved in the preparation of loan documents for their customers; and that in any case they have failsafes to stop employees editing data in any case (encryption of the data; no USBs allowed for use by company employees, etc). Despite all this there have been some isolated reports in very fringe journals that some Shinoken employees sought to duplicate loan docs to enhance incomes – though these allegations were denied by the company. Conclusion: this is a tougher issue to get a complete handle on, given the business similarities between the two companies. However, again – it has now been well over a year since the Tateru scandal broke; the company has been subject to all kinds of examination; and the most that has come out have been some isolated and unconfirmed allegations of minor impropriety. While this issue may take further time to dissipate, the fact that the stock price now essentially values the apartment business at a negative value (as we shall see) suggests that at the very least, even an orderly diminution of the development segment would create positive value for shareholders.

 

Where does this all leave us?

Let’s go back to Shinoken’s fundamentals. It should be no surprise that in the midst of all these issues, the real estate development business – which houses the core apartment development franchise as well as the condominium development business – has seen earnings fall aggressively. Essentially banks have taken a much more circumspect approach to lending against investment properties, and – even for ‘good’, trusted developers like Shinoken – this has had a large impact on business. Revenues fell 14% YoY in 1Q, and 40% YoY in 2Q (lapping tougher comps as the scandal really only first hit the business in 4Q last year). I fully expect 3Q (ending Sept) to show another large decline YoY in the development segment, though the company has guided to a return to land acquisition and inventory build-up from 3Q, as they see a pipeline of interest returning this quarter, and expect bank attitudes to gradually improve. I should emphasize, however, that I do NOT expect earnings in the apartment segment to come roaring back for at least the next 2 quarters – both because the company has likely monetized some higher margin properties in the last couple of quarters (profits held up much better than I expected); and because construction lead times on new projects are at least 6-8 months, meaning new business consummated now won’t hit the PnL until next year at earliest.

The quality of business is improving; and the stock is half the price

Why, then, am I so bullish now? Well, other than sensing the market turning before the fundamentals finally bottom, simply put, this is a much better business now than a couple of years ago, and it is available at half the price. That is to say, the deceleration in apartment business has only served to highlight how much better in absolute quality the overall business is becoming, with a much higher share of recurring revenues and earnings from management and services, and less reliance on lumpy development revenues. At the same time, the business is available at – I think – around 4x trough earnings (FY19E), or the cheapest it has ever traded, and still down 50% or so in absolute terms over the last year:

In other words, we get a much better business, at half the price of the old business, with earnings likely to start growing again from next year. Oh, and we also get a near 4% div yield while we wait (well covered by recurring earnings alone), and management’s promise to pay 20% of earnings out in future years. On my numbers, this implies a growing 5% yield from next year – a very solid carry while this business continues to compound and sentiment retraces from extremely oversold levels. And of course we are paying a fraction of book value for a long-term compounder that – even with today’s deleveraged balance sheet – is still generating 20% RoEs.

 

A quick Sum of the Parts

We can think about the cheapness on the headline earnings level (4x current year EPS and returning to growth next year); on the multiple of book or invested capital relative to its returns on equity or invested capital (as discussed earlier, at 0.7x book value for a historical compounder at 40% of book and even today generating 20% RoEs); or, more completely, we can do a sum-of-the-parts (SoTP) analysis to see exactly what we are getting when we buy Shinoken. Here is my simplified view of the latter:

It’s always important to be wary of assigning too much weight to a SoTP – especially in Japan where it is difficult to crystallize or extract the value in this way (by breaking up the parts). But I think it is instructive to see how much latent value there is here (and so how much margin of safety we have at the current price), considering the following:

• I have assumed the RE Development business is only valued at Current Assets less ALL liabilities – essentially assuming it creates no further economic value beyond what is on the balance sheet (indeed, only the current assets!) today. This seems quite punitive (after all the business is still quite profitable today);
• I value all the other businesses at the very low end of direct comps (including just a 12x multiple on the real estate management jewel, when listed REITS trade at >20x EBIT);
• I capitalize corporate overhead at a relatively penal multiple (7x EBIT) despite the inherent truth that most corporate costs relate to the core development business (which, in this exercise, is effectively being valued at adjusted book and not considered a going concern);
• and I ignore all long-term assets, optionality around the Indonesian REIT business, investments on B/S, etc as well).

Despite all these fairly punitive assumptions, it is not hard to get a value north of 2000 JPY/share, or basically 1.5x higher than where the stock trades today. Indeed even at that level, the stock would barely trade at 10x P/E on trough-ish earnings – hardly a demanding multiple, so I think this is very much a realistic mid-term target.

 

Final thoughts: real estate market risk

It would be remiss of me to write a full investment pitch about a real estate developer and manager without a few words at least about the general property market environment in Japan. However, I really don’t want to focus too much on the broader market – simply because I really don’t think it is at all central, or even that peripheral to this specific investment thesis (such has been the impact of idiosyncratic issues upon Shinoken’s cheapness today). Yes, if you buy this you are implicitly betting on the Japanese property market remaining buoyant – but again Shinoken is <1% of new housing starts in Japan and had, until recently, grown consistently in the pre- and post-Abenomics, ultra-low-rate environment so I don’t think you are necessarily beholden to current low rates as you may think. Having said all that, the attached report demonstrates a) there is still huge demand for investment properties in Japan (across all stripes, including residential); and b) that cap rates for residential real estate continue to compress. If you believe that ultra low rates persist in the medium term in Japan (a pretty safe bet given the trajectory of the last 30 years), then the broader secular risk regarding real estate in Japan are far, far down the name-specific risks related to this investment.

 

Disclosure: long Shinoken (8909.JT)