Chapter 10

Planning & Management

as of May 20, 2016

We all plan.  Planning is good.  I do not go to the food store without a list, and I do not make the list without visualizing what I intend to eat, and determining what I need that I do not have to achieve that end.  Actually, it is more common that my partner Bethanne makes the list, and I effectuate it.  Division of labor, you know.  Planning is the thinking part of doing.

Architects, for example, are planners.  Without them, would builders know what to build?  To the extent that architects are engineers, calculating loads and their distribution, they certainly are necessary.  I have found them a pain in the butt, and done many a home renovation without one.  In Portland, Oregon, I purchased a house in the west hills with a stunning view of the Tualatin Valley.  The house hung off a cliff, standing on tall stilts.  Made of wood, the stilts were rotting in the wet Portland atmosphere.  I had to replace them with steel.  Thanks a lot, architect.

One was immediately drawn to the valley side of the house, either to look out the livingroom windows, or to stand on one of several decks.  Yet there was also a deck on the street side of the house.  Who would ever sit there?  And there was hardly a dining room, only a corridor to this deck.

I decided to enclose the deck, make it my dining room, and make the corridor a vestibule to the dining room, holding hutches with flatware and dishes and the like.  I engaged the original architect to prepare drawings for the city’s inspector.  In his sketch he added a window looking out to the street.  What nonsense!  The point was to look the other way.  A window to the street would serve no function, as it would have to be covered to maintain privacy.  This is planning?  It was the very reason I didn’t consult him for design, only for a drawing.

Windows serve several functions, such as letting in light, allowing one to see out, and perhaps allowing fresh air to enter.  All of these functions can be obtained in other ways.  I built skylights into my new diningroom roof, and built a solid wall on the street side of the room, where the architect had placed a window.  Now I had light, but privacy.  What a glorious room!

The kitchen-family room of my home in Hendersonville, North Carolina, had standard eight-foot ceilings.  Walking up the mountain behind the house I saw the shape: The roof was pitched, peaked in the center.  Why did I not see that shape from the inside?  I asked my workers to make the shape of the family portion like that of the roof and, by the way, with skylights.  They tenderly took out part of the ceiling to show me the elaborate structure of 2×4 lumber.  “That holds up the roof” they said.


“No it doesn’t,” I countered.  “It holds down the ceiling.”  It is true, of course, that outside walls will be pushed out by the roof, that you do need some structure to keep them upright.  But a complicated strut every 24 inches?  Not necessary.  The new ceiling, with its depth to increase strength and accommodate insulation, would be sufficient.  I am happy to say, I was right.  Just look at this room.  Imagine the family that built it, living with a flat ceiling at the height where the wood walls end.  The architect gave them a box, where they could have had majesty and light.


One problem with planners is lack of feedback.  They plan, they build, they walk away.  Another problem is that whatever feedback they get is from the wrong people.  Some landscape architects are just beginning to get the idea that people will walk where they want to.  Two schools are emerging.  The first develops methods to prevent people from going anywhere the planner has not ordained to be the correct place.  The second plants grass, waits until paths are clearly delineated by where people walk, and then forms sidewalks to accommodate them.

The pathways determined by the people using the space is feedback.  The issue is whether the planner wants or accepts it.  In Washington, D.C., I noted this struggle for years between the Metro stop and the Department of Labor.  It was a straight shot from one to the other, but on an angle the planners had not allowed for.  Grass was worn down into a path showing where commuters wanted to walk.  The planners tried to rope the grass off.  That didn’t work.  Ultimately they placed jagged rocks along the border.  One could cross the rock barrier to get to the grass, the preferred path, but only at the risk of breaking an ankle.

I take it the planners were successful.  They may have presented a paper at some planners meeting, showing how to keep walkers in their place.  That may have been how the Hendersonville Library planners came to the same solution, to prevent patrons from “cutting the corner,” shortening their way into the entrance off the parking lot.  Some people may wonder what that pile of rocks is doing next to the entrance, but others know.  We have been put in our place.  We now walk through perfect 90º angles to enter the library.

I am not so democratic that I think the masses should always get their way.  Think what our culture would be like if that happened!  But when we plan for outdoor public space, perhaps the people that space serves should be our paramount concern.  The so-called experts don’t think so, which is why they are not experts at all.

The Textile Industry

Weaving is something I know about; and the weaving industry, also.  Remember the weaving industry?  We used to have one.  In the United States, it started in New England, for three reasons.  First, the English came to New England, including people with textile manufacturing experience and the ability to build machines.  Second, there was water power, rivers of enough substance to power water wheels.  Third, there was a dispersed agricultural community, which encouraged the making of goods from cloth at home, by hand.  The mill could weave cloth, distribute it with designs and instructions, and pick up garments or other products.  “Cut and sew” is born.  Today we would call it “distributed processing.”

Electrification allowed the mills to move south, and northern labor unions induced them to.  As the south was where cotton was grown, being close to raw materials became more important than being close to water power.  Still, the picture is not so neat.  Mississippi is where cotton is grown, and North Carolina (say, Greensboro or Charlotte) became the center of the weaving industry.  It took over a hundred years for the industry to move south, and even now there are vestiges in Rhode Island, Pennsylvania, New Jersey, possibly every state.  There are other natural fibers besides cotton.  Alpaca “farmers” in Ohio and upstate New York do their own spinning.  There may be wool from sheep in every state.  Wool from Montana is considered among the finest in the world.

Much else occurred between the industry’s move south and my first-hand knowledge of it.  We have the demise of the distributed work system and the rise of the factory not just for weaving, but for making other end-products.  With machines that harvest, clean and comb cotton; with machines that spin raw material into yarn; cut-and-sew became the most labor intensive part of the textile industry.  So it moved to other countries.  For example, Bon Worth developed a large cut and sew operation in Hendersonville.  By the start of 2011 they still would cut in Hendersonville, but then ship the cut pieces to Mexico, where they were sewn.  That is, the hand work is so much less expensive in Mexico that they can pack, ship, unpack, sew, pack, ship, and handle all the accounting that this involves, and still come out ahead.  By the end of 2011 all of that, the cutting and the sewing, was moved to China.

In the 1980s, with the introduction of computers, came the resuscitation of the Jacquard weaving industry.  It is that industry, in particular, that became centralized around Charlotte.  That is where Dornier, Staubli, Bonas, Somet—the manufacturers of looms and Jacquard heads—still have corporate offices.  It is where Benninger and Mayer and Hacoba, the warper manufacturers, located; and Suzuki when Asian technology started to be competitive.  Many have now merged (Somet is found as part of Itema, Benninger as part of Mayer, for example). But the remnants remain in North Carolina.

American textile manufacturing killed itself.  As onerous and expensive as governmental regulations and taxes are, and as real as the wage difference between American workers and, say, workers in Mexico or Bangladesh or Thailand is, the force driving textile manufacturing from the United States was decisions of the management of American textile companies.  Although some of those decisions were correct, some of the basic ones were not.  We are paying the price for inexpert textile manufacturing management.

2-DraperLoomsI will not dwell on American manufacturers of weaving equipment.  Technical change, however, is interesting.  A major production issue was how to keep machines weaving when the source of weft yarn ran out.  That source, traditionally, was a shuttle, which carried its yarn supply with it, wound on to a “pirn,” through the “shed” opening of the warp.  When that supply was exhausted, a new pirn was inserted into the shuttle.  Draper, the premier American loom manufacturer, devised a clever automation, ejecting the old pirn, dropping a new one into the shuttle.  Although I have seen such carousels with different yarn colors, by and large weaving was relegated to one weft color using this system.

Meanwhile, the rapier weft system was devised in Europe. The rapier selected from as many as twelve possible weft yarns, brought a single yarn across through the shed (the raised and not-raised ends of the warp), then returned to get another one. A double rapier system, in which another rapier came from the other end and grabbed the weft yarn in the middle—cutting the insertion time in half—produced speeds many times greater than the shuttle system. Draper failed, and no standard industrial weaving machine is made in the USA today.1

That looms are not made in the United States is part of a broader loss of capital goods producing industry.  As rapier, and then air-jet looms were available from European manufacturers, why did the fabric-making industry also decline?

If one wants to explore this phenomenon, I suggest that The Peter Principle might help.2 The Peter Principle is not only succinct and understandable, it is most likely correct.  The principle asks how people advance in organizations.  The answer is, they do well at one level and thereby are promoted to the next.  People advance by succeeding until they reach a level at which they do not succeed, or at least do not stand out.  By not advancing from that level, they populate that level, and so high levels of management—the Peter Principle goes—become inhabited by people whose performance is mediocre.  There they have the most authority they ever get and, being there through failure, generally exercise it badly.

These might be the people designated to select experts to move the corporation or government agency ahead.  This was the level at which Sears chose the corporate officer to supervise the design and building of the Sears Tower in Chicago.   It might explain the terrible decisions made by British Petroleum that led to the 2010 oil spill in the Gulf of Mexico.

The Peter Principle assumes that the better performers are promoted to positions where they perform other tasks than those they used to do. Will they do these new tasks as well?  As is well-known in statistics, there is another explanation why the top echelons of a corporation may not contain the best talent.  That explanation is called “regression to the mean.”  It says that in our measurement of quality of prior performance, there is actual performance, and there is error or perhaps luck.  There is a part of the measure that is determined not by the quality of the person being measured, but by other circumstances.  Of two candidates, one will “look” better than the other, and be promoted.  Is he actually the better candidate, or the one who happens to score highest on the measures used by those making the decision, who may not be experts at that task?  Some bad promotions, in this explanation, occur not because the tasks change, but because the measurement of candidate quality was faulty..

It does not matter which explanation is correct.  It does not even seem to matter whether the promotion is vertical (within the same organization) or lateral.  Worldcom, Time-Warner, Hewlett-Packard, J. C. Penney provide examples of destructive top officers.  Microsoft badly selected Bill Gates’ successor.  Steve Jobs was dismissed from Apple, Inc., in favor of John Sculley. Can you imagine that?  Sculley wanted to put more resources into the Apple II.  Jobs wanted the resources to go to the Macintosh.  Jobs, we know, was correct.  The Board of Directors was not.

Although it cannot be universally true that higher management is less competent than lower management, the people who get to be top decision makers in management are not necessarily—not likely—the best people for those jobs.  Just like politicians, like school board members, the characteristics that get them there are the characteristics necessary to be chosen, but not the characteristics they need to do the work.  And so they fail, and are sent off to pasture with multi-million dollar retirement packages.

The textile industry was created by people who knew how to make cloth.  As it settled into the south, and individual mills merged into corporations, the question became where to get management.  At first, making cloth seemed not to be the important skill.  Then the question became how to make money.  Managers were recruited to meet that criterion.  Not experts at making cloth, not experts at managing people, but so-called experts-to-be at making money: MBA graduates.  People for whom every aspect of a mill is a cell in a spreadsheet.  Thus, the industry was doomed.

I am a mill owner.  I weave yarn into cloth, and make cloth into products.  Although you might think I revel in the failure of my competition, the opposite is true: Their failures are hurting me in real ways.  Thinking that we were competitors, that one would succeed over the other’s failure, is part of the mentality that has led to this decline.  Indeed, I will conclude that there was no “industry” at all, in the sense of competitors recognizing their common interests.  There is a professional tennis tour (or golf tour, or bowling, or pool, etc.), for example, because professional tennis players, although competitors, have common interests that are enhanced through cooperation.  The computer industry sets standards so all computers can communicate with each other and with peripheral equipment.3 Textile manufacturing never got to that level.  The “expert” textile managers were not good at what they did.  They were particularly inept at understanding where the firm ended, the industry began.  And sdo there never really was a textile industry, just many textile producers.

A Primer On Weaving

I know, you did not sign up for this course.  But ask yourself, how many academic types that you have come across have ever actually done something?  I do not mean done something like write a book.  I mean engaged in commerce.  Started a manufacturing or distributing or retail business.  And yet they lecture at you about how business operates.  And look at the management consultants who pontificate about best practices, coming up with new best practices every few years, because once they have written their book, they have to find something new to say.

One of the great problems with so-called experts is that they do not do that which they talk about.  They are experts in knowing others, in credentials, in writing anecdotes from which they are unable correctly to generalize.  As weaving is my example, let’s get a small feel for weaving as an industry.  Let me tell you, it’s pretty damned interesting.

Weaving is done in all cultures.  It is inherently rectangular and binary.  It is rectangular because yarn goes in two right-angled directions, let’s say up-down and left-right.  Up-down we will call the warp, which is set up before weaving starts.  Weaving, then, involves the lifting up of some of the warp threads (called “ends”), and passing weft threads (these days called “filler”) through the “shed” formed by the some-up and some-down warp.  I called it “binary” because each warp thread is either up or down in the formation of this shed.

Most weaving is done with the aid of a loom, which holds (and advances) the warp, makes the shed, and passes (or allows the passing of) the weft yarn.  There is a final “beating” stage that tightens up the cloth, and then it is usually rolled up.  Some cloth is then finished off the loom, but much of it is not processed further.

3-backstrapOn the right is a simple hand loom.  Some such a loom is common throughout the world, and provides the cloth that dresses much of the world’s population.  Beautifully, I must say.

Below on the left you see the beginning of industry, a stationary loom with two “harnesses” whose job is to lift up some of the warp threads.  We see it from the back, from the warp, and so do not see the cloth it produces.  With a two harness loom you invariably lift all the odd ends with one harness, the even ends with the other.  Lifting each harness in turn (by stomping down on a pedal), passing the weft through the shed, you are making “plain weave.”  The weft yarn on such a loom is inside a wooden “shuttle,” as discussed above.  In hand weaving, the shuttle is thrown from one side to the other. 

The first step towards automation was hitting the shuttle with a mechanical arm, to propel it across the shed. The shuttle would have to start and end in a box, of which there could be several on each side (one on top of another). The next step, therefore, was to determine which box would contain the shuttle to be hit, and which box would receive it. This determination still exists in CAD software as “box motion,” although there no longer are such boxes.

The arm was originally actuated by the weaver pulling on a rope. This was called a “fly shuttle,” and did increase weaving speed several fold. The next step in automation used the same approach, but motorized the passing of the shuttle, and the beating of the reed into the previous cloth.4


When automated, this harness-type of loom is called a dobby.  The mechanisms that lift the warp ends are called frames.  If the weft is inserted by shuttle, it is also called a shuttle loom.

Weaving technology has advanced on two fronts.  The Jacquard mechanism requires a separate Jacquard head, mounted above the loom, capable of lifting each thread separately.  Thus it allows much more complex designs  The Jacquard mechanism cannot respond as quickly as dobby harnesses, and so is intrinsically slower.  In weaving, slow means expensive, and indeed Jacquard woven fabrics are more expensive than dobby fabrics, partly because the machines are more expensive, partly because the cloth takes longer to make.  Fewer than ten percent of woven goods are Jacquard woven.

I discussed weft insertion above.  The second front on which technology increased the speed of weaving was freeing the weft yarn from the mechanism that transported it from one side to the other.  Be it by rapier, air jet or water jet, the mechanism is fed by two packages of yarn, the end of one tied to the start of the other.  When the first package has been exhausted, weft is automatically taken from the second.  I call this the “infinite weft” system, a brilliant increase in the speed of weaving.

In some industries technology reduces cost, increasing demand for the product and for people to make it. The weaving industry is beyond that point.  In the 1970s and 1980s, “expertise” was lost in both capital goods, the making of looms, and in manufacturing fabric.

5-LoomsAtOrioleThis is a picture of some of the looms in my mill, in 2008.  The orange cords hang down from the Jacquard heads.  We are seeing, on the left, a loom that uses each hook in the Jacquard head to lift two warp ends.  A design takes up half the weaving width, and then repeats.  The loom to the right has no repeats, although it is difficult to tell at this angle.  There are also two other Jacquard looms in the background.

This view shows the rear of the looms, and thus the warps.  Making warps is a big deal, requiring a separate machine in another room.  One could skip the warping process by putting a creel holding hundreds (even thousands) of “packages” of yarn, extending far back from the actual weaving action.  I have seen some polyester warps on creels, but not natural fibers.

Few mills leave as much space between looms as I do.  If different looms are weaving different fibers, the lint from one will contaminate the fabric in another.  There are many such quality issues in textiles (for example, the speed at which one runs the looms), but if the object is to make money, not fabric, they are ignored.

Nor will you see another mill with natural light coming from glass block along all four walls.  Our designers stand at the loom during testing.  They have good light, and can make quick decisions.  Many mill designers never enter the weave room.  Their samples are delivered to them in the design area.  The separation of functions and the physical separation of people induce an attitude of “it’s their problem.”  Every experienced loom operator has, as a first reaction to a weaving problem, “it’s the designer’s fault.”  When you put the designer and loom operator together, they no longer think of “fault,” they think of “problem” and “solution.”

Most of you will never see another mill.  Or an automobile factory, or a dye house, or a steel foundry.  We are brought up divorced from how things are made, and how social relations in the workplace influence how things are made.  We can have this separation from production because we are so wealthy, because most of us will never grow anything, much less see that raw material transformed into something else.  That wealth, which translates for many people into leisure time (that is, time one can do with what he wants, as opposed to time required for maintaining one’s family), is surely a remarkable advance.  Yet it is not without cost.  One of those costs is that we rely more and more on “experts” to get things done for us, assuming they do it, in some sense, “right.”  If they are not expert, what they do is wrong.  If people never experience truly expert quality—or if they rationally determine that quality is not worth the price, to them—then over time the quality of our products will decline. It is not for me to say whether that is good or bad, but it redefines the word “expert” to relate to the production process (“cheap and fast”) rather than to the quality of the end product.

The Scalamandre Story

Now at least you have a picture of textiles as an industrial product, requiring expensive machinery and skilled workers, and managers who know how to put them together.  You also have my word for it—not proof—that managers, selected for their “expertise” at making money, failed to do that, or failed to do that within this country.  To illustrate,  here is a textile story.   Scalamandre is one of the great names in American weaving.  Franco Scalamandre immigrated to the United States in 1924.  He started Scalamandre Silks in 1929.  Although some of their products have disappointed their most effete customers (the White House and the Getty Museum in Los Angeles both went to France for their fine silk woven products), and I find their modern (computer) designs badly conceived, they nonetheless historically have produced some beautiful fabrics.  Here, for example—on the loom on which it was being made—is what appears to be a simple stripe design.

6-ScalamandreStripesAlthough it appears that these stripes are of different material, they are all made from the same natural silk.  The stripes arise from the light reflected differently by different weaves.  For such weaves to work, because they use different amounts of warp, they have to be on separate warp beams.  And they are.  Below is a picture of the rear of this loom.

There are three warp beams.  Silk ends, interleaved with Kraft paper, are meticulously placed so the three beams feed one seamless fabric.  The fabric design is sophisticated in its simplicity, featuring the sheen of the silk in some stripes, not others.  It might well go back to Franco’s time, to Franco himself.  Fine, also, is the workmanship that was able to produce this beautiful fabric in the first decade of the 21st century in the United States.

No weaving this difficult is done in this country any longer.  This was Scalamandre’s Jacquard Loom Number 6,  in Gaffney, South Carolina, where Scalamandre had relocated in the late 1990s.  They left a comfortable building they owned on Long Island to rent space in a a cavernous, corrugated steel, no-character building, owned by Coker International, a used textile equipment dealer.  Production was half-owned by another company, Metropolitan.  It produced as Scalmet. They occupied a small corner of the building that served mostly as a Coker warehouse.

This is the business model of used equipment dealers.  As weaving mills went under, dealerswould acquire their property, as well as their equipment.  They paid for this by re-selling the equipment to surviving mills.  In effect, they “purchased” large buildings on large tracts of land, for no net money.  They would then profit, they thought, from selling the land.7-ScalamandreLoom6

Coker had recently purchased the Dan River Mills facility just south of Greenville, South Carolina.  It was moving its warehousing there, which left the Gaffney facility—in which Scalamandre was the only tenant—mostly empty.  To sell weaving equipment from the warehouse in Gaffney would save Coker the cost of moving it to Greenville.  That was why I was invited there to look around.  That is why these photos exist.

The situation inside the building was painfully obvious.  Scalmet had not paid rent in months.  Coker had a mostly empty building, which was quickly becoming valueless as industry around Charlotte was contracting.  The famous Scalamandre, now reduced to half of a two syllable name, was on its way out of business, at least out of the business of manufacturing its branded products.

Not that you can tell from the Scalamandre web site, which says not a word about where they manufacture.  Scalamandre fabrics—all of them—are woven under contract by a company in India.  This great American manufacturer neither got good enough to equal foreign production, nor contracted with foreign companies who could live up to its reputation. In 2009 the third generation of the Scalamandre family sold the entire business.  This is the modern American textile story:  Scalamandre became a marketing company, with a New York show room, living off of its reputation (and its raw material, silk), but selling cheaply made, inferior products.  And so it failed in its last task, sales.  Franco’s grandchildren were no longer interested in making fabrics.  Their sole interest was in making money.  At which they failed.

Scalmet’s demise left their employees, who had moved from New York to stay with the company, unemployed.  I hired three of them, and engaged another as a contractor to thread a particularly difficult loom.  I assumed that if these people could make the warp for their Loom 6, and weave that cloth, they were of the highest skill in the industry.  I thought my getting Scalamandre employees was a coup.  Multiple experts, all at once.  I was so wrong!

Neither these workers, nor two other experienced crews I hired, knew anything about manufacturing quality fabric.  They did well-enough the tasks that they had been given, tasks they did over and over again.  Punch a clock, perform tasks, punch a clock, go home.  This is who domestic weaving workers were.  Why?

When I taught “Radical Economics” at Harvard, one of the points I made about capitalist production was that management exercised “control” by not sharing its “secrets” with labor.  Rather than encourage workers to learn, they encouraged them to remain ignorant of the firm’s whole operation.  Since the 1920s, manufacturing was divided into small tasks.  Do your task, it is management’s job to see that it is a proper and productive use of your time.  Rather than encourage them to view the mill as theirs, and therefore to contribute to its success, managers made clear who really owned the mill, and you’re damned lucky to have a job here.  Yes, the workers have bad attitudes that make them not so good workers; but that attitude has come from management. From very bad management.

I also taught that, although limited to particular aspects of production, workers had knowledge management needed.  Workers know particular details, such as the proper speed at which to run each loom, about which management has no clue.  So that when management becomes oriented toward making money, instead of making fabric, they wonder if the loom’s speed can be increased.  They do not ask workers about the wisdom of doing so.  They just instruct the workers to do it.

I have found this picture to correctly describe declining mills.  Management not only does not know enough to share its thinking with workers, it does not know enough to ask workers to share their ideas and knowledge, or give them a reason to do so.  Seeing  manufacturing as an adversarial process, management could not operate the cooperative process it would have to be.  By the time I met the Scalamandre crew—decent folks all—they were ruined as workers.  They were clock punchers and task performers, not weavers.

Not that low wages and lack of environmental regulations are irrelevant.  It is cheaper to produce abroad, but cheap enough to be worth the communication and transportation cost?  And the product inconsistencies?  I do not think so.  The problem lies with management.

Henry, the “expert” we hired to thread our 19,200 end loom, turned out not only to be a prima donna, but to make a serious error.  He missed 16 ends maybe 800 ends in from the finish.  Someone would have to take those 800 ends out, insert the 16 missed ends, and then re-insert the 800 where they belonged.  Henry would not do it.  Henry admits no error.  Someone else must have sabotaged him.  Henry does the job perfectly.  Just ask him.

To hell with “experts” like that.  He has never worked for me since, not because of his error, but because of his attitude.

What should have been the most expert crew in all of Jacquard weaving, at least in this country, had neither the skill nor the mentality to join a start-up, to get equipment working, to produce fabric.  I have great sympathy for the union movement, which was necessary in the early 20th century to counter management mistreatment.  But unions have essentially the same view of workers as does management—that they are all the same, and not really responsible for the product.  Both sides see no distinction between worker and machine, except that whatever tasks workers do, machines will do in the future.  There is a lot of truth to that projection, as long as workers are considered to be task performers, not thinkers.  That is why workers fight automation, which they see as displacing them, not as making their jobs easier.

In our mill, Bethanne and I define the outcomes.  We set the goal and the structure. We define positions as outcome responsibilities. Our employees structure the tasks required to get the job done. This is the opposite of the traditional mill, in which management and engineers devise the production tasks, and who (or what) will do them. That is the sense in which they are all cells in a spreadsheet—task cells. The MBAs “know” what tasks need to be performed. Whether humans or machines will do them is determined by their relative cost.

Once one defines the future as machines substituting for humans, one sets the present structure as occupied by humans acting like machines.  Until recently, workers could be reprogrammed more easily.  Like machines, they would be maintained at some minimal level.  Some would get broken and be discarded.  Management would substitute one for the other whenever it became clear that it was economical to do so.  Workers added no value to machines—they were simply alternatives.  That is why the famous John Henry tried to beat a steam-powered drill, working with only his muscles and a pick.  They were both doing the same thing.  The only question was which would produce output (broken rock) more cheaply.

It was inevitable that management would put as much effort as possible—restricted only by cost—to inventing machines that can do labor’s tasks.  As long as managers and workers saw themselves as competing for the owner’s dollar—competing among each other, and with machines—then that is how they acted.  Why would owners see them any other way?

Thus the biggest failure of management is the attitude of the workers.  If the workers have no stake in the enterprise, not necessarily legal ownership, but some stake and say in how it operates, to what standards, then why wouldn’t they see as their role to do only the tasks assigned—no more—and collect their pay?  Machines will not volunteer to do extra tasks, and as humans and machines are essentially the same, under this view, why should the humans?

That attitude served the Scalamandre crew well enough in a going concern, but for a startup like mine, it was inappropriate.  Scalamandre’s workers had been so beaten into their roles that they had no ability to exercise their skills on their own. The stripes made on Loom 6 proved that there were skills in that crew, but could they be put to work fruitfully for me, or were they only silk stripe-making skills? Unless told precisely what to do, and supervised while doing it, the old Scalamandre crew did nothing.

Franco Scalamandre’s daughter ran the business well.  She had two sons, Mark and Bob Bitter.  I have no sense that they worked their way up, starting by sweeping the floors.  They inherited management positions, not the right to earn management positions.  Their mother “managed” them poorly, and they, in turn, managed the business poorly.5

That may be the “why” in that particular situation.  Let’s look for a larger “why,” a more general explanation of the decline of American manufacturing, as exemplified by the loss of its textile industry.

Was There A Golden Age of Management?

Jack Buffington has this notion of “classical management:”

The role of classical management is to introduce the invention and introduction of technology concurrently with the destruction of dying processes and industries to ensure that capital and labor resources are appropriated efficiently for economic growth.6

I’m sure this means something to someone, but the reason I quote it is because Buffington says there was such a thing as classical management.

It started around 1908, developed strongly in the 1920s and 30s, flourished in the 1940s, and then started to slack off in the 1950s and 1960s.

Buffington writes emotionally about this now past golden age.  Classical management introduces the introduction, eh?  What it does, he says, is optimally allocate resources, giving labor its appropriate share, and stimulating economic growth.  Except for correctly seeing that current management is unproductive, this is a fantasy.  There are too many motivations and undefined terms (stimulate macro-economic growth, pay labor “fairly,” allocate resources “optimally”) to be a system.

He praises Henry Ford for both vertical integration and raising workers’ wages, while noting that Dell Computers, trying to imitate Ford’s production concepts, could not successfully do so.  That is not surprising, as even Ford Motor Company could not sustain its growth, just as Microsoft has not been able to.  Perhaps Ford’s success, like that of General Motors, was consolidation and monopolization of a new technology—it came more in corralling capital and marketing than in production efficiency.  We know that Microsoft is one of the world’s worst managed companies (look at its products!), but by being owner of the IBM operating system, it created a niche that it exploits successfully to this day.

Buffington also praises Frederick Taylor, if not the inventor then certainly the master of time and motion studies.  Taylor had three key concepts about work: First, a job was a set of tasks and those tasks could be written as instructions.  Second, performance of those instructions could be timed.  Third, one could devise an incentive system to get workers to fulfill more instructions (more tasks) in a given period of time.  That incentive system could be more pay, but it was never as much additional pay as additional work.  Management success, to Taylor, was getting a worker to be at least 20 percent more productive for at most 10 percent more reward.

Buffington also feels kindly toward Andrew Carnegie, finding it contradictory that he left actual management of his steel firm to Henry Clay Frick, who hired thugs to beat striking workers into submission.  There is nothing contradictory, or golden, or “optimal” (optimal to whom?) about any of these characters, all of whom understood that monopoly power was not exercised only against competitors, but also against workers.  One has to hope that “the invisible hand,” the notion that individuals working for their own self interest will produce benefits for all, has merit.  Selfishness, or greed, certainly has produced a lot of stuff, and some of that stuff surely has made many lives better.  But that it works consistently, or that it is the product of a field called “management,” are both surely false.  Even if consumers benefit, workers have historically been abused and unappreciated.  Workers have by and large both hated and feared management.  They came to work reluctantly.  How could this be a good system?

Managers have authority to make decisions.  In theory they do so for the good of owners, although it is clear they mostly do so for themselves.  Management would work better if managers were required to be owners, and if they took most of their rewards from that ownership.  One of the first things managers would do, in that circumstance, would be to reduce the number of managers in the firm.  But as long as they can siphon funds from owners and workers to themselves, they have no incentive to reduce their numbers.

John L. Lewis, as President of the United Mine Workers of America for forty years, also employed thugs when it suited his desire to maintain control, but he had more of a grand vision of the allocation of resources than any coal mine manager.  Lewis’ concept was that it was labor’s responsibility to gain a share of productivity increases.  Management would voluntarily do no such sharing.  Similarly, long-shoremen understood that containerization made loading ships more efficient, that is, profitable.  They organized to get a share of that profit, not to, prevent it.

Lewis drove miners’ wages up, understanding that it would reduce union membership.  He convinced union members to give up job security for pay.  Changing the relative prices forced management to mechanize, making the miners’ work situation better and more productive.  To credit management for any of this is to fail to see the struggle that continues to this day, between mine-owner profit and mine-worker safety and reward for service.  It is no surprise that in describing the golden era of classical management,  Buffington does not mention coal mining or any other extractive industry, or farming, or shipping.

Buffington’s is nonetheless an important if misguided view, because he does stress the role of management in the failure of business in the later part of the 20th century and beyond.  I am not sure “management” was ever enlightened, or operated on principles that were good for itself or anyone else.  But I agree with Buffington that it is none of that now.  If we are looking for “experts” at running businesses—and if we expect those experts to be good for the economy in general, as well as for their firms—they are hard to find.  I think the issue is not so much the unavailability of individual managerial talent, as the lack of a progressive system in which that talent can work.

Why Is There Management, Anyway?

To believe economists, the great achievement of economic theory is the recognition of price as the result of many individual actions, and its function as the allocator of resources.  That is, the masses, by buying and selling, generate the signals that drive industry and consumption, that tell us what to produce, and how, and to some extent what to consume, and where.  There is no expert.  There is no need for an expert.  “Planning” on this scale is doomed.  The market regulates for the public good.

This is what we are told.  There is much to admire about the concept, but to leave every planning decision to “the market” would be a disaster.  Those whose solution to everything is “let the market decide” either control the market, or are just mouthing words they learned in a class that might as well have been called religion as whatever it was called.  Experts, they are not.

The very existence of managers tells us that many decisions are not made by “the market,” but by people.  The existence of schools of management indicate that there might be better and worse decisions being made by these managers, independent of market forces.  Whether these schools induce better decisions again begs the question, better for whom?  But they clearly indicate that leaving all decisions to “the market” is a bad idea.

There have been several lines of thought about the economy that explore what happens if “the market” does not quite work.  One of those was the discussion of “imperfect competition,” mostly in the 1930s, featuring Joan Robinson.  It was easy to see that the pre-conditions for “optimal” resource allocation under perfect competition—numerous fully informed buyers, numerous sellers with transparent prices, etc.—did not exist.  What should economic and legal policy be in such a situation?  I do not think that Robinson and others successfully answered that question.7  Perfect competition has nice mathematical properties and so leads easily to pronouncements by economists that may have no value in the real world.  Imperfect competition, in contrast, leads to case-by-case evaluation and ultimately a subjective judgment, no doubt made by “experts.”

In 1956 there was a step forward, in which two economists showed that when the pre-conditions for perfect competition were absent, a lot of the conclusions one would draw from a perfect competition model were not just questionable, but demonstrably wrong.8  However, this work did not advance the cause of economists as experts in the economy—its conclusions were not so much what to do in a particular situation, but what not to do—and was drowned out by the clamor for “market” solutions led by economists who apparently did not get the point.

Paul Samuelson formalized the difference between private goods and public goods, the former but not the latter being “best” provided by perfect competition.9  A good is private if your consumption of it prevents someone else’s consumption of it.  An apple, for example.  We do not know in the abstract what the cost would be to produce an apple, or how many people would pay that cost.  The best way to find out is to let producers and consumers both compete.

If you build a road, however, my use of that road will not interfere with your use.  If we leave road building to the private market, too few (relative to some “optimal” quantity) will be built.  Public goods have to be provided by the public, financed often by means that are unrelated to the individual benefits received.  Indeed, one of the benefits of roads leading to your house is that emergency services can use them to help you, whether they are ever called upon to do so.  Those same roads provide that same service—or potential service—to your neighbors.  Even if your particular road serves few of you, and can easily be privately built (as in a development), that is not true of the city roads and highways by which goods and services get close to you.  No one argues against public provision of roads.  Or, if someone does, do not listen to him.

Then there are goods that lie in between public and private, such as multi-room houses.  We think your parents own the house, and their consumption prevents enjoyment of its benefits by anyone else.  If you had your own room growing up, then your parents did not have it.  Although a single unit, a home has private and public sectors.  You use the livingroom together, but the bathroom separately.

Then you are unemployed and ask if you can move back in.  There may be some loss of benefits to your parents, but in the larger picture more people consume it than did before.  Many people can consume an apple, by cutting it into pieces.  That is not a successful model for allowing many people to use a kitchen.  Now that you and your parents share a kitchen, you will have to figure out how to use it to everyone’s satisfaction.  The market is not the solution to all of our problems; not in fact, and not in theory.

For the purposes of this chapter, perhaps the most interesting alternative line of thought was developed by R.  H.  Coase (and others).  Coase had been told by his instructors that the brilliance of “neo-classical” economics was the allocation power of price, price that resulted from unplanned behavior, from the individual actions of producers and consumers.   Yet, when he looked at the world around him, he found

that most resources in a modern economic system are employed within firms and therefore how these resources are used depends on administrative decisions and not directly on the market.10

The instructions given to workers within a firm are not automatic translations of “market” prices.  Indeed, there could be no innovation if they were.  There has to be planning, and management, and organization outside of the “market,” for most things to get done.  Coase not only realized that “perfect competition” or “let the market do it” was not the solution to all problems, he quickly realized that government had to set out a structure within which decisions were made within the firm.  And thus the field of Law and Economics was born.

Another way to see Coase’s contribution is through the notion of transaction costs.  The traditional model assumes that all parties are fully informed.  In fact, getting information, especially getting correct and timely information, is expensive.  The solution when there are transaction costs may not be the same as when there are no such costs.  But those costs do exist, which is why the “perfect competition” model, let the “market” decide, often is wrong.

We all look to minimize transaction costs.  One method is branding—stay with the brand you know.  Another solution is relationships.  I do not bid out every carpentry or construction task I have.  I just call Jim.  He understands what I want, gets it done excellently, and gives me an honest bill.  The hardest thing about management is finding the employees or contractors who will think with you, then ahead of you, and have the skills and judgment and attitude to get the job done right.  Ask any manager of a small business why he is successful, and he will tell you that it is because of the staff he has assembled.  Ask any manager of a textile mill why he is successful, and then wait as he tells you many things, but not about the quality of his staff.  Few textile mills are successful, because management is not expert.  They do not understand the knowledge embedded in their workers, or how to use it.

Corporate Management

Managerial “expertise” is identified in two places: In management schools and in the popular press.  What are managers taught?  What models are managers supposed to emulate?  What tools are they supposed to use?  Obviously this is too large a topic for this book.  Sorry.  I’m no management expert, you know.  But I have run a 39 person statistical research firm, and currently run my small manufacturing entity, my weaving mill.  Unlike professors at business schools or in economics departments, and unlike management-advice gurus, I have first hand experience as an entrepreneur.  I have put my money and my life on the line.  I think I know something about management, particularly how badly it has been done, how destructive of American industry it has been.  But, not being a management professor, I am not considered an expert.

Managers of big companies get to make big decisions.  They get to effectuate those big decisions when their Board of Directors agrees.  I have mentioned Apple’s determination that Steve Jobs was more of a liability than an asset.  They forced him out, and floundered until he came back.  They made one of the worst decisions in corporate history, yet the company stayed alive.

Another contender for “worst decision” is Time Warner’s purchase of America On Line (AOL).  Or did AOL purchase Time-Warner, and why would we care?  I don’t find the idea of combining a print and television empire with an internet site totally stupid.  For example, Yahoo has a finance-knowledgeable staff that could probably be utilized in forms other than its finance search site, so that, at least if one were a specialist print organization (like Investor’s Business Daily), there might be some synergies.  Perhaps the Time Warner-AOL deal has scared everyone off, so that no one wants Yahoo whole, or perhaps everyone else knew something that Time Warner did not, namely that they did not have the management skill to make the merger work.

Carly Fiorina, as CEO of Hewlett-Packard, had to fight her board to purchase Compaq.  I thought she was right, but apparently she did not have the skill to handle the merger, even though she had the insight to instigate it.  Hewlet Packard—HPQ, as it became—found another CEO to put the companies together.  He was not exactly a success.  When he left, they considered selling off their computer division, which is to say, to unwind the deal they had spent years and billions of dollars putting together.  Investors left HPQ, because management obviously did not know what it was doing.  Then they hired Meg Whitman, who revived the idea of splitting the company, and has done it.  These CEOs, these managers, have been paid millions of dollars a year to recirculate the same ideas.  There seems to be little expertise at these lofty levels, and no relationship between quality and reward.

The apocryphal story about how management counts is of Thomas J. Watson understanding that IBM would make business machines, not just tabulating machines, as it did (under a different name) when he became its head.  Having a vision of what your company does is surely important, but even Steve Jobs, when he returned to Apple, probably did not have a vision that included an iphone.  His vision, rather, was of a company that innovated, designed, saw where technology was going for the consumer.  His vision was to lead, but where to lead would evolve over time.  That, some people say, was Fiorina’s problem:  She could see HP acquiring Compaq, but once they had, she had no idea what to do with it.

The achievement of great managers is first to set a vision—think of Walt Disney as an example—and second to understand that such a vision cannot be achieved by one person.  The next management task is establishing the organization that buys into the vision, contributes towards it; that recognizes that every worker may have an insight that contributes; that establishes lines of communication and genuine respect for what each individual has to say.  There is no such vision in the textile industry, which continues to spend little on research and development that might correct the overall decline in quality.  More and more fabrics are made from artificial fibers, because they are less expensive than natural fibers.  After they have been woven, these fabrics require extensive treatment to come close to the feel of real cotton, or wool, or linen.  But they are still less expensive.

Most designs are just copies from other manufacturers.  My mill, The Oriole Mill, which does innovate, has been knocked off by others who, using cheaper materials, produce inferior but less expensive products.  That is just how the industry works—or doesn’t.

The great failure of American manufacturing has been the notion that managers and workers are adversaries, and that workers are a temporary phenomenon.  Now they can be replaced by other workers, and soon they will be replaced by machines.  If there is any single occupation, title, function that has led to the sad state of American production in the second decade of the twenty-first century, it is “manager.”  You know, the ones who make bad decisions and then make all the money.  The people who, in the media, are called “experts.”


  1. AVL makes a wooden “compudobby” loom which does allow shuttle selection. It weaves slower than the Drapers, requires stopping the loom to change the pirn when its yarn runs out, and is not dependable enough to be used in a real mill. Fletcher Industries produces one-of-a-kind high quality shuttle looms for specialty purposes, such as weaving medical material that will be inserted into the body. Mageba, in Germany, also produces high quality shuttle-based machines, also for specialty use. European manufacturers have moved on to air jet and water jet technologies that quadruple the speed of rapiers.
  2. Laurence J. Peter and Raymond Hull, The Peter Principle, Morrow (1969)
  3. In the early days of computing, each printer used its own language to turn information from the computer into print.  If you lost the driver information—which came, with the printer, on a floppy disk—you were in a heap of trouble.  After DOS, that function was taken over by the operating system.
  4. The operator pulls a handle above his/her head, which pulls a rod that bangs the back of the shuttle.  The shuttle is on only one side, but the operator need not care.  There must be an ampty box on the other side. So the operator pulls the handle, and rods bang at boxes on both sides. The shuttle, once hit, travels to the other side.
  5. “The third generation is a notoriously vulnerable juncture for a successful family business” writes Lauren Collins, “Bread Winner,” The New Yorker December 3, 2012, page 78 (quote at page 83).  Collins describes Appolonia Poil_ne, who successfully manages the bread bakery started by her grandfather.
  6. Jack Buffington, The Death of Management, Praeger (2009) at 36.  Next quotation at 3.
  7. See Joan Robinson, The Economics of Imperfect Competition, MacMillan (1933), 2d ed., 1969.
  8. R.G. Lipsey & Kelvin Lancaster (1956‑1957).  “The General Theory of Second Best,” The Review of Economic Studies, 24(1), pp. 11–32. Reprinted in Robert E. Kuenne, ed.,  Readings in Social Welfare: Theory and Policy (2000), pp.48‑72.
  9. Paul A. Samuelson, “The Pure Theory of Public Expenditure,” 36 Review of Economics and Statistics 4:387 (1954).
  10. R.  H.  Coase, “The 1991 Alfred Nobel Memorial Prize Lecture in Economic Sciences,” printed as “The Institutional Structure of Production,” in Essays On Economics and Economists, University of Chicago Press (1994) at 5-6.  I have slightly altered this quotation to make a complete sentence.