Sunday, July 7, 2019

Review of Tragedy & Challenge: An Inside View of UK Engineering's Decline by Tom Brown

It has already been a year since we did a "full review" of a book on the blog. We have come to rely heavily on the abbreviated "note" format (e.g. Q1). However, we had enough notes and thoughts about Tragedy & Challenge: An Inside View of UK Engineering's Decline that a "full review" is warranted.

It is by a Brit named Tom Brown, about the decline - collapse - of engineering ("metal bashing") in the U.K. that took place from WWII to present. For example, Britain's car manufacturers are gone, with either the brands bought out by foreign firms or else, like Austin or Morris, just gone. Naturally this means that the U.K. lost the many smaller firms that supplied components, as well.

The most interesting part of the book is the first fifty pages contrasting the first two engineering shops from the beginning of his career, where he had very hands-on experiences managing production. He started out working on the shop floor and mentions a philosophy of not asking people to do anything that he would not do, and not trying to manage anything that he has not done himself.

(Remember one of the key concepts from our 2018 book reviews: people who have not done the job they are asking someone else to do are unable to tell if their employees are lying when they say they cannot do something. The way they deal with this is to just demand an outcome and then see what happens. The type of person who is an employee does not understand this tactic and desperately tries to do the impossible.)

In "Seeing How to Do It Wrong," he is working for a U.K.-owned firm in the U.K. called Scottish Stamping and Engineering. One of their problems was union labor that was so angry or crazy that they were happy to sabotage their employers and prevent profitability. But many of the problems were also caused by management. For example, the firm would have obsolete, broken-down equipment that would slow down productivity when it broke. Replacing these items would eliminate bottlenecks at the plant and consequently have very high return on the capital invested. (See The Goal from Q1 reviews on bottlenecks.)

However, the broken down items were fully depreciated, so replacing them would have two certain effects on the financial statements: an increase in net property, plant, and equipment (and total assets), and an increase in depreciation expense over the new life of the machine. This had the potential to lower the Return on Net Assets (RONA) metric because of the higher denominator. (Note that the investment ought to result in a net increase in profit, driving up return if the idea was sound, but this is uncertain and his managers seemed very pessimistic with the result that the investments like this were not made.) Despite unwillingness to authorize these capital expenditures, his management was complacent about higher operating expenditures, like paying overtime when these breakdowns would otherwise prevent deliveries to customers. He noticed an absence of key performance indicators relating to quality, productivity, and job punctuality. His conclusion based on that experience is that "companies that won't invest inevitably rot away after a few years of fictitious prosperity." (You can also consider this an "anti-EBITDA" book; a metric that Brown calls the "milking ratio".)

In "Discovering How to Do It Right," he moves on to work for a U.K.-owned firm in South Tyrol (the German speaking part of northern Italy) at a ball bearings plant which was German speaking and had Germanic engineering culture and labor relations - which includes a positive attitude towards work. The employees were educated for manufacturing jobs in technical high schools and apprenticeships. The sales people were engineers who understood the technology and could speak the buyers' language of engineering. But one of the key competitive advantages was superior process engineering:

"[I]n such a small niche the range of specialist manufacturing equipment available to buy on the market was very limited. We rebuilt our presses in our maintenance department, which gave us the enormous advantage of being able to build in all the know-how and ideas which had been generated in the works, so that our rebuilt machines were literally much better than anything available new (and it taught me how very important this kind of production engineering and process development is, creating intellectual property of a type that is very hard to copy and impossible to buy)."
Thing like that are where profit margins come from. Too many investors, especially right now, take profit margins to be permanent features of businesses. But it is difficult to maintain these heroic levels of accomplishment, which would be why the profit margins revert. (Plus the high profits attract competition.) Nobody is going to manage with that level of intensity unless they are properly incentivized with upside or young and irrationally dedicated like Brown was.

The German ways were superior, and Germany beat the U.K. in manufacturing despite having the burden of east German reunification and lacking the bonanza of the U.K.'s North Sea oil. Brown says that this is because people in the U.K. with power and influence over manufacturing had no operating or engineering experience, just financial, and in contrast "the one thing Germans don't invest in is financial engineering."

The Boeing 737 MAX mistake has been in the news recently, and the parallels between the problems at Boeing now and the U.K. manufacturing firms before they collapsed are eye-opening. In Matt Stoller's blog about monopolies, he recently wrote,
In 2001, a Boeing employee named L. Hart Smith published a paper criticizing the business strategy behind offshoring production, noting that vital engineering tasks were being done in ways that seemed less costly but would end up destroying the company. He was quickly proved right.
The L. Hart Smith paper is worth reading - here is a good section:
The first issue to be examined, is precisely what is out-sourced and what is inevitably retained. The superficial perspective might be that every internal activity that used to be related to a task that has been out-sourced is no longer necessary. Even that is not true but, worse, it fails to acknowledge all of the new internal tasks that had not previously existed. To add insult to injury, contemporary accounting practices do not allow these unavoidable additional costs to be billed against that particular item of work – because it is no longer identified as an in-house task – so these charges are allocated instead as overhead to any remaining in-house work. This misrepresentation of true costs furthers the illusion that outside production is cheaper than anything done inside, building the pressure to ship even more work offsite, until there isn’t any left. The irony of this situation is that it is so easy to understand in the extreme. Suppose that a manufacturer had succeeded in out-sourcing all of the work that it wished to isolate from the preferred task of systems integrator. The unallocatable costs from the huge amount of out-sourced work will now appear as overhead on the few remaining tasks, like sales and product support, confirming that these were now even less profitable than manufacturing had been when the spiral began!

“What are all of these additional tasks?”, one might well ask. The first is the need to write a specification for the product, which must be more complete and precise than would be needed for in-house production, for which omissions, refinements, and improvements could have been accommodated without the need for costly legal discussions. One must ask the question as to where the skills for writing such specifications will come from if there is no continued in-house production from which to learn. The next is the obvious need for additional transportation costs associated with off-site production and the increased total span time this inevitably causes. These could be avoided by having the supplier co-locate his factory alongside the final assembly site, but only at the cost of a new factory and the perceived higher wages that were to be reduced by removing the work in the first place. No, the additional transportation costs are the least expensive of the options, once the decision to “off-load” the work has been taken. A major source of unplanned costs associated with out-sourcing has been subassemblies that cannot be fitted together because all of the drawing tolerances had been consumed earlier in the fabrication process. As a minimum, this has entailed considerable out-of-position work, a lot of rework, delays for replacement parts, and recriminations as the source of any error is uncovered. Whenever it is a drawing error or omission, added costs are incurred in changing the contract. And whenever it was a tool supplied to the subcontractor or built by him and bought-off by the prime manufacturer’s inspectors with an undetected error, the costs of fixing the mistake were even greater. These, and other, problems would be diminished tremendously if only internal organizations need to be involved. In order to minimize these potential problems, it is necessary for the prime contractor to provide on-site quality, supplier-management, and sometimes technical support. If this is not done, the performance of the prime manufacturer can never exceed the capabilities of the least proficient of the suppliers. These costs do not vanish merely because the work itself is out-of-sight. 
The paper goes on to talk about Boeing using outsourcing to drive improvement in the Return on Net Asset Ratio - the same problem that Tom Brown observed at the U.K. owned firms.
Return on Net Assets (RONA), barely qualifies as a sub-optimum solution. Superficially, it seems to be very appropriate. Who wouldn’t want to make the most from all of the assets an organization possessed? It would make an excellent goal, by any standards. Only when it is converted into a performance metric do its devastating effects become apparent. As a ratio, it can be enhanced by either increasing the numerator or decreasing the denominator. Each activity can be related or totally unrelated. Far too many additional issues are involved. But unless one studies all of the cause-and-effect relationships created by changing either numerator or denominator, one cannot possibly tell if a course-of action is good or bad for a company, no matter whether RONA is being increased or decreased. Conversely, if one were to apply RONA unthinkingly, once could easily convince oneself that decreasing assets was always a good thing, because doing so would always increase RONA – assuming that changes in assets being employed never had an effect on the costs of production!

It is the author’s belief that this process, and others like it, is what has led the aircraft industry into a state of excessive out-sourcing. It is obvious that doing all of the work in-house would require a greater supply of capital equipment than subcontracting it all outside. That is not the issue. The first issue is whether or not an incremental change in capital facilities or other assets results in a beneficial or undesirable change in revenues and associated profit. There is absolutely no guarantee that a reduction in assets will ever result in an increase in RONA, let alone an increase in profits. That is the great fallacy about this metric. It is applied by far to many with the authority to do so as if reductions in assets are always to be preferred over increases in capital spending. The second issue is whether or not that same incremental change in net assets results in any collateral or downstream consequences that had not been included in the initial cost/benefit assessment. This brings us to the next great fallacy about the concept of RONA. No allowance is made for the fact that company revenues and/or profits can vary greatly for other reasons, while the level of assets might remain unchanged. Using the ratio of two sometimes totally unrelated variables as a basis for strategic business decisions simply doesn’t make sense!

RONA might be a worthy overall goal, but as a performance metric, it is usually meaningless – and a serious threat to the survival of any organization on which it has been inflicted. The problem with RONA is its superficiality and its plausibility if one does not think too seriously about it. A typical stock-market analyst lacks the deep understanding of what makes different companies tick and why their profitability cannot always be assessed by the same metrics. The misapplication of ANY of the financial assessment tools can do, and has done, great harm to the livelihoods of far too many people.
That paper was written in 2001, about three decades after the U.K engineering decline in Brown's book. It turns out that the previous Boeing CEO who made the decision to just "upgrade" the 737 was an MBA with no experience in aviation. (He has never had any type of pilot certificate!) He did have a career at GE, including in jet engines, but his first job was in "brand management" at Procter & Gamble.

So what are the common elements at Boeing and the U.K. firms, which you would not find at German or Japanese owned firms? There seem to be two: company managers with no technical or scientific knowledge and weak, ignorant shareholders who also lack technical knowledge. The dearth of information that the owners of public companies have about them reminds us of something in the book Panic by Andy Redleaf:
What modern capital markets do very well is raise large amounts of capital from a broad base of investors who are persuaded to give their money to perfect strangers with precious little idea of what these fortunate recipients are going to do with it. In order to keep the money coming in under such admittedly odd circumstances, liquidity and the universal, instantaneous "price discovery" that financial markets offer with a glance at a computer screen are essential. The public investor, knowing so little about what he is buying, must be able to tell himself he can get that money back (or what's left of it) pretty much whenever he likes. […] Public securities markets, and especially equity and derivative markets, are bad markets because their knowledge base is thin (at least compared to the sum of what could be known about the underlying companies if shareholders were allowed to know it, or inclined to learn it).
The index investing trend - the desire for a free lunch through diversification - has led to fund management by generalist investors with pedigree, connections, and confidence but not technical knowledge. They operate based on social proof (see Theranos). Brown says that the professional U.K. fund managers invested in his companies could not understand anything technical and would not engage with anything besides financial results. He said that they would invest in offerings brought to them by the sell-side with very little diligence, "taking it on trust from the brokers that they were being given an inside track on a good thing." He said it was "extraordinary what obvious questions the fund managers [would] fail to ask."

As a result of this, not only are securities prices obviously inefficient, but so is society-wide allocation of capital. Brown points out that the investors in his companies had "enormous portfolios of shares in up to 200 companies" - something which no entrepreneur would ever have, and which left them with no ability or inclination to act as owners of the companies.

In contrast the best engineering firms were those of the German Mittelstand, which are mostly family owned. He says "many German companies would not even know what their weighted average cost  of capital was, and would only consider financing costs if specific financing was necessary for a project."

1 comment:

Anonymous said...

Great review and extremely interesting read! A fascinating story that seems to be almost universally relevant to so many Western companies that are being managed in precisely this fashion by the plague of MBA graduates. No question that many of them don't understand their industry (viz. the idea of the interchangeability of management and irrelevance of specific industry knowledge). What I find fascinating, novel, and so exquisitely horrifying is the idea that they don't even understand the implications of their own tools (financial metrics).

I had never really considered how financial metrics, if poorly applied, could be fatal for a business. It's logical once one things about it - but since one is dealing with abstractions (as all financial metrics are) it's easy to overlook the translation of their impact into the real world. It's no different than using the wrong gauge to monitor a machine: if oil temperature is the critical metric, but someone is monitoring (and optimizing) air pressure instead, the results can be fatal for the equipment.

Another palette of bogus metrics: diversity and inclusion. The company where I am currently working (a pharmaceutical giant) is going all-in on D&I initiatives. They're having a big kick-off this week for the program with a highly-paid female consultant from PWC leading the charge. I looked at her LinkedIn profile: no value added to any real core interest of any business she's ever been involved in. But making big $$$ managing the (likely multi-million dollar) global roll-out of a program whose sole purpose at the end of the day is to dilute the true talent of the company with people who fit some demographic wish list. So these MBA types have now completed the circle: it started with the idea that a (generic) manager could - with the right tool-box (of financial metrics) manage any company, regardless of what they do. Now these people think even talent is interchangeable (replace white-guy engineer with black-woman engineer = no loss of skill + D&I points). The problem of course is that at the end of the day, somebody has to actually do something transformative to something in order to actually produce revenue. And skill - I mean real talent in a professional that enables an engineer to channel knowledge, experience, and creativity into revenue-generating innovation - is rarer than people think (and then one must take into account motivation: the brightest engineer in the world will underperform if unchallenged, neglected, or unappreciated). So the question is: how many bean-counters and D&I placeholders (as well as the staff managing and administering these programs) can the actual, revenue-producing talent of a company actually carry and for how long? Be interesting to know if there's a way to actually model that.