Description

Elevators are toll roads in the sky.  Like escalators, they are very localized, physical monopolies on people movement.  The elevator and escalator (“E&E”) industry is an attractive one for the large Original Equipment Manufacturers (“OEMs”) that have dominated the industry for more than a century.  They get paid in advance for new installations, sourcing commodity components globally, assembling and installing finished product locally at bespoke prices, with a good shot at a lifetime of easily renewable maintenance contracts often mandated by local regulations.  No wonder the bulk of industry economics accrues to a few chosen ones.

Toll roads: funded by customers

Kone (€ millions)

Why now? 

The best time to buy elevator stocks is at cheap valuations or at the start of a cyclical upswing in new installations.  Neither of those conditions applies here.  The absolute valuation is high, despite being lower than the market.  And 30% of Kone’s 2016 global sales were in China, which is almost certainly past a major cyclical peak:

So why now?  Like you, I must write-up fresh longs for VIC even when I don’t find many I like.  This idea is as far from a table-pounding as I have so far submitted.  Nevertheless, if I am right about industry dynamics and business quality, buying this stock should still do ok when compared to an even more richly priced index.  I hope that something of my write-up helps other members when confronted by better odds one day, or that it provides industry context for Jt1882’s recent excellent write-up of a smaller, much cheaper name: https://valueinvestorsclub.com/idea/Golden_Friends_Elevator/139958

Threat of new entrants: How does a wannabe get a piece of the action?

The minimum entry price is a reputable brand build on years of minimal defect journeys, and economies of scale in both manufacturing and maintenance.  But for a second tier player to get promoted from marginal to outstanding economics, there are four main strategies.  (Spoiler alert: the first three are rarely successful on their own; the final one has worked spectacularly well, so far, for Kone).

1.  Build a better mousetrap.  Elevator trade show participants often claim that they want to see new technology, because it is important to the industry.  Nevertheless there is not much difference between a high quality elevator car from 2017 and 1997.  Similar to automobiles, the underlying mechanics have been very slow to change.  And unlike autos, fuel and batteries are unlikely to change much for decades.  So a great leap forward in the underlying technology is a difficult entry point for new competitors.  Big advances have been slow in coming, but exclusively from the incumbents since the modern industry was founded on the ability to prevent accidents by surviving a rope break in 1852 with Otis’ safety elevator. The second major innovation took more than 50 years, when Otis introduced the gearless traction elevator in 1903.  The third major design change took almost 100 years, when in 1996, Kone developed the machine-room-less elevator: particularly well suited to its core European residential market.  Along the way, new technologies have been relatively inexpensively adapted to E&E products in the fields of textiles (longer, lighter ropes replacing the steel cords whose weight limits elevator shaft height) electronics, communications and sensors.  Industry R&D rarely exceeds 1.5% of sales.  None of these innovations or adaptations have required a new entrant.  Elon Musk need not apply.

2.  Compete on price.  Cheapo Elevators Inc. faces the classic new entrants’ dilemma.  They cannot pitch new installation work to customers without an extensive track record of safety, which can only be acquired by successfully winning many new installations.  Few companies therefore get to quote on new installation work.  But many companies attempt to scale into a fully-fledged E&E business by offering discounted maintenance services.  Field engineers trained by the OEMs frequently launch independent firms to meet demand from price-sensitive building managers.  However, achieving scale beyond a local maintenance portfolio is extremely difficult.  Evidence for this is the number of such potential new entrants that get acquired each year by the OEMs (Kone alone acquires dozens of these each year for $2-10 million each throughout Europe and America).  The OEMs exploit this reality to knockout irritating local competition and simultaneously improve the density of their maintenance economies of scale.

3.  Pick up the crumbs that OEMs drop.  Some smaller players such as Kleeman adopt this strategy, building out a full product offering, and targeting customized jobs.  But despite profitable orders, their growth is limited by existing capital: the growth dilemma faced by successful new entrants.   Without a dominant position in a market, they cannot force new customers to fund a large proportion of future growth by supplying working capital, so returns on capital do not exceed mid single digits.  And until returns on capital comfortably exceed cost of growth capital, those growth opportunities remain unexploited.  Access to cheap capital from customers ultimately separates the truly wonderful E&E businesses from the also-rans. 

4.  “Go West East young man” – exploit new geographic markets.  I would say that Kone only become a terrific business, as evidenced by sustainably delivering returns on equity above 25%, from 2006 onwards.  Not coincidentally, this was also when Kone’s global market share reached above single digits, climbing up to today’s 19-20%.

Founded in 1910, Kone therefore took 96 years to finally become a great business.  During that first century, three generations of the Herlin family had applied the first three of the strategies above, with the grit and consistency required to build any global business.  But still, results were insufficient, as significant market share gains were elusive.  The transformative opportunity came in the 1990’s by spotting a massive potential new market – China – that was evolving fast.  Crucially, incumbents persistently underestimated its potential, even as it rapidly became the largest E&E market in the world.      

Antti Herlin, aged 40 in 1996, took over the CEO reins from his hard driving father, Pekka Herlin.  Today Antti is Kone’s chairman, owns 22% of the business and controls the voting stock.  One of his first strategic decisions in 1998 was to build a large new factory in Kunshan, China.

“Wow, this is a once in a lifetime kind of thing.”  (Jim Chanos)

At the time that Antti was building his first factory in China, the North American E&E installed base was about 600,000 units, significantly bigger than China’s 350,000 - a mere speck of the 6 million units worldwide.  But Kone needed a big new geographic market to open up in order to leverage new machine-room-less technology, its willingness to accept lower margins and to seek out business that other incumbents neglected.  Mature, predictably profitable markets offered few openings to improve market position.

Seven years later in 2002, Chinese new installation unit demand had doubled.  Three years later in 2005 it had doubled again.  Four years later in 2009 it had doubled again.  This was the year of Chanos’ famous insight:

"The story is internally now one of our great stories.

A real estate analyst was addressing the partners and he said: 'Currently there's 5.6 billion square meters of high rises in China under construction. Half residential, half office space.' And I thought for a second and I said: 'No, you've gotten the American, rest of the world metrics wrong. You must mean 5.6 billion square feet. Because 5.6 billion square meters is roughly 60 billion square feet.'

And my analyst looked at me sort of terrified. He was a young analyst at the time. He said: 'I know. I double-checked. It's 5.6 billion square meters.' And I thought for a second and I said: 'Well if half of that's office space, that's roughly 30 billion square feet of office space. And that’s a five foot by five foot office cubicle for every man, woman and child in China.'

And that's when we all looked at each other and our jaws dropped.  Realized, wow, this is a once in a lifetime kind of thing, where this whole country is in effect building itself out in a very short period of time.”  (source: Business Insider).

Around this time, many businesses and investors, long uneasy with the ideology behind Chinese’s apparent relentless economic growth, swung 180 degrees in their macro enthusiasm for China, and concluded a Chinese crash was inevitable.  Kone, from tiny Finland with its 5 million people who had for generations learned how to prosper alongside a dominant (and one-time sovereign) neighbor: communist Soviet Union, blindly marched on, building Chinese market share while others pulled back.  Kone finally reached #1 market position in Chinese new equipment in 2012.

Between 2009 and 2013, new installation units in China had doubled again.  Within 2 years, new installations had peaked (there is a lag between real estate new starts and E&E installations), and Kone is now navigating its third year of declining new installations.

It is certainly possible that the most bearish forecasts on China play out, and the collapse will be epic.  But I also think that simplistic macro predictions on China will prove…simplistic.  Just like there is not one Chinese real estate market, but hundreds, the once in a lifetime build out of China’s real estate should leave a complex, enduring legacy of opportunities for E&E companies with global scale.  In year three of what might yet become a multi-year, massive decline in new equipment, China’s market is currently far from a zero for Kone.  This year – 8 years after Chanos’ call – total new units delivered in China will likely still be 20x the size of the entire North American market. 

Even if no new units get sold there ever again, China’s current installed base represents 30% of installed E&E equipment in operation globally, compared to just 8% for all of North America.  Unless lots of Chinese urban dwellers head back to the countryside leaving bricked up elevators, or businesses tenants, metro station passengers and airport users vacate commercial and infrastructure buildings en masse, the opportunity of just maintaining the already-installed 4.2 million units exceeds the next four largest countries (3.7 million combined for US, Spain, Italy and Germany).  With Kone’s #1 market share in China new installations, and a 50-60% trailing conversion rate with room to improve in years ahead, there should be plenty of profitable opportunities to adapt to an unfolding reality.

Kone’s Chinese maintenance business is still tiny, and faces two obstacles:

Market structure, partly due to immature regulatory framework and poor compliance.  Major OEMs only maintain 25% of China’s installed base; 75% are maintained by 7,000 independent service providers, often the building managers themselves.  Kone’s conversion rates – the percentage of their new installations that, after a grace period of up to 3 years in China, eventually convert to Kone managed maintenance contracts – are very low for them at 50%.  Otis China is at just 35% (up from 24% conversion in 2010), and even its 50% target for 2020 will be much lower than its typical achievement globally.  This is a major drag on industry profitability for the OEMs, who consider it abnormal to their experience elsewhere.  But it also makes the new installation market less competitive than it otherwise could be, since new installations are not used as loss leaders for ‘inevitable’ fat maintenance contracts.

Technological threats to maintenance monopolies.  Innovation outside of the industry has made sensor technology price competitive, which means that building managers have full performance data transparency and will potentially be able to predict, or at least closely monitor maintenance issues themselves.  This will erode the role of the maintenance “man in a van”, by increasing customer scrutiny of actual work performed under maintenance contracts.  There is already some evidence in European maintenance that new technology is eroding aspects of the industry’s economics.

But there are two reasons to believe that Kone will be a long-term winner in Chinese maintenance:

Economies of scale still wins.  A long-term view I think attaches a high probability that maintenance markets will organize themselves according to economies of scale.   For example, despite all the investment in online education tools, still the most efficient method of training new engineers is exposure to multiple repair scenarios.  The engineer that repairs the most elevators is most productive, and therefore lowest cost.  Mom and pop elevator maintenance is not a sustainable end-point for the largest elevator market in the world.  Personally I am still very careful about my elevator use in China.  But the day should come when policy and local industry matures to the point that horrific accidents are minimized, through the efficient allocation of maintenance resources, to the benefit of a few large service providers.  I think Kone’s largest installed base in China (and current joint #1 maintenance market share there) means that it should stand a good chance of being a major beneficiary of this long-term trend.

China’s transition to relative E&E safety will not be overnight, but will still be quicker than most Western countries’ experience.  E&E accidents hit social media fast.  Every tragic accident that occurs, rightly attracting significant attention, should improve Kone’s conversion rate.  This trend should remain in tact even into a Kyle Bass-style collapse.  Maintenance price pressure is fierce in social housing, especially in some of the Tier 3 and Tier 4 ghost cities that could yet make the margin erosion of the Spain’s elevator industry 2009-2017 look like a walk in the park.  But 2 million or 50% of China’s installed base by units, and more by value, are commercial and infrastructure.  Maintenance decisions by large state owned enterprises are not just economic, and purely commercial building managers face rapidly improving quality expectations by users.  Regulations have started to become more stringent in recent years.

Conclusions so far

·      This industry is very hard to enter, when entry means earning decent returns on capital.

·      It took the once in a lifetime build out of China for Kone, a century-old, global company, to start playing in the big leagues.  It was precisely the uncertainty about Chinese growth prospects that created the competitive opportunity for them.  Consensus future profitable outcomes would have been much harder to capture.

·      Experience over the first century and half is that it just takes a very long time to build sufficient scale and reputation in this industry.  Long-term some global competitors should also emerge from this, e.g. Canny Elevator, but so far there is no globally scaled E&E equivalent like Lenovo, Xiaomi or Huawei who have emerged in other industries.

I.         Threat of new entrants: Low

II.         Threat of Substitutes:

·      New installations: very, very low

·      The lack of alternatives raises switching costs.  Product failure costs are huge relative to price. 

                                      i.     For most end-users living or working in high-rise buildings, aside from a short-lived seasonal health drive, the stairs are not an option.  A building manager that skimps on elevator quality or maintenance loses tenants.

                                    ii.     The manager of a subway system will think twice before rejecting a service renewal quote if escalator defects have been low.  A single escalator malfunction could create a bottleneck and quickly cripple a whole subway network, even causing fatalities as passengers exit a train, deep underground, with few viable alternative routes to the surface.

                                   iii.     Airports are big buyers of escalators and elevators.  Downtime of one node in their network creates delays for passengers and crew, cascading through aircraft schedules and costing a lot more money than an apparently expensive maintenance renewal.

·      This improves business quality but also reduces risk.  Technological obsolescence threatens elevators and escalators less than most industries I consider. 

·      Maintenance: high

                                      i.     Increased data transparency to customers.  Mobile apps can now show very specific details of jobs performed, a recent innovation that is coinciding with declining maintenance profitability in the biggest European market.  Maintenance portfolios in Europe have all seen margin deterioration in recent years.  Otis saw revenues per unit drop 10% compared to a 3% increase in costs 2010-2016.  Though this could be all attributed to aggressive cost management from real estate managers hurting during prolonged weak macros in key markets such as Spain, it could also be more sinister.  Greater customer bargaining power over maintenance contracts seems likely to result from greater data transparency (“I paid you maintenance contracts for 50 residential units last month, but you did not need to attend 48 of them even once!”).  Peace of mind is a great selling point, so anything which achieves this through technology rather than a padded maintenance contract is bad news for incumbents. For this industry to continue to earn non-commodity returns on capital for many more decades, maintenance providers will need to convert more of their field engineers’ time from unproductive lazing around and looking busy to the customer, to actual value added.  Kone and others are investing in this through better use of technology and investment in education.  Otis, with the biggest maintenance workforce of long-tenured European mechanics, consider only 50% of hours worked actually add value.  Their strategic response is to deploy 31,000 smart productivity tools from the Otis app store.   Expect fewer, better paid field engineers in future.

III.         Rivalry among existing firms

·      New installations:             high

                                      i.     But can vary based on firm’s macro outlooks (e.g. Schindler’s late enthusiasm for China)

·      Maintenance: varies by geography from low to high

                                      i.     Aided by regulatory barriers to entry in some markets.  Rivalry is structurally lower for maintenance contracts in France – where monthly maintenance site visits are mandatory, than in Sweden – where the law requires zero maintenance visits.

                                    ii.     Otis: 1.9 million units under maintenance worldwide

                                   iii.     Schindler: 1.4 million

                                   iv.     Thyssen Krupp: 1.2 million

                                    v.     Kone: 1.1 million.  Kone maintain not only their own equipment but also other manufacturers.  So rivalry in maintenance does exist.

IV.         Bargaining power of buyers:         low

·      A fragmented customer base (millions of real estate managers and owners) buying from a few large companies is a dynamic that most investors appreciate.

V.         Bargaining power of suppliers:     low

a.     Fragmented supplier base, typically operating on commodity margins.  Otis has 22,000 suppliers, 80% of whom sell less than $10,000.

Additional Risks

·      Global real estate crash.  Although this is a better business than most, it will still look pretty bad during a crash.

·      Fat technology tails: improvements in remote monitoring, maintenance prediction, AI and ultimately hardware design allows self-maintaining technology.

·      Chinese policy decisions.

Catalyst

·      China's "inevitable" real estate crash is more nuanced than anticipated.

·      Maintenance conversion rates improve in China

·      Global infrastructure spend increases.