“New World Order”: Foreign Affairs

The July/August 2104 issue of Foreign Affairs contains an interesting conceptual economics article titled “New World Order.”  It’s written by three professors–Erik Brynjolfsson (MIT) , Andrew McAfee (MIT) and Michael Spence (NYU)–and outlines what the authors believe are the major long-term trends influencing global employment and economic growth.  I’m not sure I agree 100%, but I think it’s a reasonable roadmap to start with.

Here’s what the article says:

the past

Globalization has allowed companies to exploit wide wage differentials between countries by moving production from high-cost labor markets close to consumers to low labor cost areas in the developing world.  Former manufacturing workers in high-cost areas enter the service sector to seek employment, depressing wages there.

This period is now ending, as relative wage differentials have narrowed.

now

Relative labor costs are at the point where manufacturing plant location is determined by other factors.  These include:  transportation cost, turnaround time for new orders and required finished goods inventory.  This implies that manufacturing can be located closer to the end uses it serves.  However, globally higher labor costs also imply that new factories will be much more highly mechanized than before.  Robots replace humans.

As a result, wage growth will remain unusually subdued.

the future 

Although returns to capital have avoided the erosion that has befallen labor over the past generation, this situation won’t last.  Long-lived physical capital is being replaced by software (note:  the majority of investment spending done by US companies is already on software).

Software doesn’t have either the total cost or the permanence of capital invested in physical things.  Software can be moved, it can be duplicated at virtually zero extra expense.  To the extent that software replaces physical capital as a competitive differentiator, it makes the latter obsolete.  It, in turn, can be made obsolete by the innovative activity of a small number of clever coders.

Therefore, the authors conclude, returns on invested capital (especially physical capital) are already beginning to enter secular decline.

Where will future high returns be found?

…in the innovative activity of talented, well-educated entrepreneurs.

education

This brings us to a major problem the US faces.  It’s the relative slippage of the domestic education system vs. the rest of the world, and an increased emphasis on rote learning (No Child Left Behind?).

The trio dodge this politically charged issue–they do observe that there’s a direction relationship between the quality of a community’s schools and the affluence of its citizens–by asserting that online learning will come to the rescue.  A child stuck in a weak school system will, they think, be able to in a sense “home-school” himself to acquire the skills he needs to succeed in the future they envision.

my take

What I find most interesting is the presumed speed at which the authors seem to think transition will occur.

–Is it possible that we’ve reached the point where there’s no available low-cost labor left in the world?  If so, this is a dood news/bad news story for low-skill workers.  On the one hand, downward wage pressure will stop.  On the other, robotization is going to take place at warp speed, making it harder to find a job.

Relocation of factories will also have implications for transportation companies, warehousing and even the amount of raw materials tied up in company inventories.

–Does software begin to undermine hardware so quickly?  Certainly this the case with online retailing and strip malls.  But how much wider is this model applicable?

–If the key to future growth is young entrepreneurs, then the sooner we as investors reject the Baby Boom and embrace Millennials the better.  This, I think, is the safest way to benefit in the stock market if the New World Order thesis proves correct.

 

 

rent vs. buy: financing and Solarcity (SCTY)

My California son, Brendan, got me interested in SCTY a while ago.  SCTY rents solar panels that generate electricity to individuals and to companies.

From an analytic point of view, it’s a complex and interesting firm.  It may also eventually turn out to be an important component of the nation’s power generation.  But it’s by at least a mile the riskiest stock I own (both Brendan and I hold small positions).  For instance, SCTY is a JOBS Act company , so the financials it has published to date aren’t ready for prime time.  Its business is heavily dependent on government subsidies of one type or another–and they’re shrinking.  It’s part of–but not at the heart of–the Elon Musk empire.  So holding it runs counter to the time-honored rule that you have your money as close as possible to where the entrepreneur has his–in this case, that would be Tesla, I think.

In this post, I want to use SCTY to  illustrate that in the rental model, a company can have an immense call for capital in advance of the business generating much revenue.  This can pose a significant risk.

Here goes:

First, note that I’m making the numbers simple (read:  pretty much making them up) and that there are many, many more moving parts to what SCTY does than I’m going to write about here.  But I think what I do say gets to the essence of the matter.

the business basics

1.  Look at a typical rooftop solar panel array that SCTY installs on a single family house.

–the panels cost $10,000 to build and install

–they have a 30-year life

–the homeowner signs a 20-year contract to pay $50 a month to rent them.

2.  In this industry, there’s some urgency to get panels installed on rooftops, at the very least because once someone has signed a 20-year contract, he’s not going to switch to another provider.  So the first mover has a key advantage.

financing new customers

Suppose SCTY installed panel arrays on 50,000 rooftops last year and wants to install another 100,000 this year.  What do the money flows look like?

Well, $30 million is coming in in rental income from last year’s installs.  But this year’s installation program will require $1 billion!! in capital to complete.  Where is this money going to come from?

In many senses, SCTY is a startup.  It doesn’t have deep pockets or an existing cash-generating business to use to fund the panels.  So raising $1 billion, and presumably more than that next year, is a formidable obstacle.

my point

That’s the point of this post–that the upfront capital committment in a rental business–especially involving physical stuff–can be very large.  From a financial point of view, some rental/service companies aren’t that much different from owning, say, an oil tanker, a steel blast furnace or a cement plant.  Not so glamorous if you look at them this way.

what SCTY does

The SCTY solution?  …the installed solar arrays are each sort of like a bond, that is, they pay a fixed amount of money each month for twenty years.  At the end of that period, the array still has ten years of useful life and therefore hopefully a substantial residual value.  If you package up a big bunch of them, the result doesn’t look that different from a collection of car loans or home mortgages.  In other words, the bundle is a security that you can sell to institutional investors who are looking for fixed income investments.  That’s a bare-bones version of what SCTY does.  Of course, it doesn’t hurt that SCTY is run by a financial entrepreneur.  Not every solar panel company is going to have the size or credibility to do this.

 

 

 

 

 

rent vs. buy: examples

Olympus Optical of Japan (yes, the huge-derivative-speculation-losses, put-out-a-hit-on-the-new-chairman Olympus) was the first company to open my eyes to the value of the rental model vs. selling an item.  The company makes endoscopes, the computer plus coated fiber optic cable devices used in colonoscopies and endoscopies.  Olympus initially used a razor/razor blade model to sell these devices around the world.  It sold the computer devices at slightly above cost.  The fiber optic cables were supposed to be the razor blades, being replaced at regular intervals with new ones, generating high profits.

Olympus didn’t make much money from endoscopes, however.  Physicians generally refused to buy replacement fiber optic cables, even when Olympus salesmen told them they risked having the cables break apart in patients’ bodies.

So Olympus tried an experiment in the US.  It switched to a rental model.

Let’s say an endoscope kit sold for $60,000 (a number I made up).  If so, the new idea was to rent the units, throwing replacement cables to avoid safety problems, for $1,000 a month.  Because $1,000 a month was easier for a doctor to stomach than $60,000 upfront, more doctors signed up for the machines.  In addition, because Olympus was collecting rent for each machine over something like a ten-year useful life, reported profits skyrocketed.  Yes, this is partly a question of accounting technique (more about this tomorrow), but the amount of money Olympus ultimately collected for each machine was double what it had before.

Anixter, the wire and cable company.  This was one of the first companies I covered in my career as an analyst.  Back then, Anixter’s main business was industrial wires and cables.  It ran a national system of warehouses.

The Anixter salesman would call on a customer, ask how much wire and cable inventory a company had–usually a lot more than anyone realized– and offer to buy it all on the spot.  Anixter would guarantee to meet all the company’s wire and cable needs from Anixter warehouses.  Outsourcing to Anixter would mean the customer could repurpose its warehouse space, lay off or reassign the three guys who dealt with the inventory, and free up, say, $10 million the firm had lying around in wire and cable stock.

The manager who shifted the company from owning its own inventory to working with Anixter would be a hero in the eyes of top management.  People couldn’t sign on the dotted line fast enough.

At the same time, although apparently not many clients realized this initially, there’s no going back from a decision like this.  If you’ve taken a victory lap for creating $10 million in cash out of thin air, as well as saving $300,000 in annual expense, you can’t subsequently return to the board to say you need money to build a new warehouse, hire new employees–and, by the way, you need another $10 million (or more) to fund inventory.

As well, in the case of Adobe, there’s no place to go back to.  As the company put it, ADBE has burned the boats.  It no longer sells its media products.  It only rents them.

Electronic Arts  In the early days of MMOGs, I was at an analyst meeting for ERTS.  Someone asked how many users the company had for its MMOG.  The then-CFO, long since retired, said he didn’t know.  All he knew was that the company collected $10 a month from 180,000 credit cards.

I took this to mean that the company had a significant number of people who were renting the game but never using it.  This isn’t necessarily a good situation.  You’d prefer that people love your service so much that they’re heavily engaged every day.  On the other hand, the no-show users are pure profit.

Tomorrow:  the Achilles heal of rental, the upfront capital needed to get going.

rent vs. buy: why rent a product instead of selling it?

Adobe (ADBE) used to sell physical copies of a given edition of its Creative Suite of products to individuals or small businesses for $2600 apiece.  Now it rents the same thing as Creative Cloud for $50 a month.  In 2012, selling physical copies (let’s ignore the other cloud-based tools ADBE sells–the big change is in its media tools), ADBE made $1.66 a share in profit and had $2.24 in cash flow.  This year, having gone totally digital the company says it will have earnings of around $.30 a share and will generate, I think, $1 or so in cash flow.

How can this be a good deal?  It takes over four years of rental income to generate the same revenue that a sale would do all at once.  In addition, in a world where interest rates were back to normal, present value considerations make the rental stream worth less than cash in hand today.

So why switch?

I can think of four reasons:

pricing umbrella   $2600 for Creative Suite, or $700 for Photoshop alone, leaves the door wide open for a competitor to enter the market with a lower-priced product–even a shareware entry–that does more or less the same thing as an ADBE product.

piracy  I’ve seen bootleg copies of Creative Suite on Craigslist for $100.  Yes, they’re illegal and, yes, maybe they won’t all work forever, but still the price difference is enormous!  Back when I was following Microsoft carefully–which is over a decade ago–that company thought that almost half of the copies of its Office suite being used by small- or mid-sized companies were stolen.  Because the rental model matches the cost of the software more closely with the potential buyer’s cash flow, stealing the software becomes much harder to justify.  If it’s all on the cloud, it’s impossible for most people to do.

upgrades (or lack thereof)  Before I signed up for the cloud version of Photoshop, I was using a version (CS5) that was several years old.  I’m sure there are individuals and businesses using much older versions.  Same general argument as for piracy–using outdated tools become much less worthwhile.

selling direct  Delivering Creative Cloud products through downloads eliminates the commissions paid to distributors of physical copies.  It also eliminates the expense of making the physical copies, but I think that’s a minor expense (the box and shrink-wrap are probably the largest cost elements).

 

ADBE thinks it will make $2 a share in 2015 and $3 a share in 2016 because of switching to the cloud for its media tools.  I’m not sure these number make the stock cheap at today’s price (I have a small position and would be a buyer at lower levels), assuming they come in as ADBE anticipates.  But I’m convinced that the piracy thing is real and that the incremental cost of selling an extra copy is as close to zero as you can get.  Also, once you start using the better tools it’s highly unlikely you’re going to go back.  You’ve probably thrown out the disks anyway.

Therefore, there’s at least a shot that number s are better than that.

But in this post, my main point is that the rental model is an extremely powerful one.

Examples tomorrow–Anixter, Olympus and EA.