The future of the Internet in milling

by Emily Wilson
Share This:

We all know how fast the world moves and how rapidly developments that capture all-encompassing attention often vanish. But hardly anything rivals the coming and, yes, going of the way in which American millers are using the Internet to originate wheat.

At the start of this year, the odds favored that by the time of this convention a half-year later I would be telling you how many American millers were conducting a large part of their wheat procurement by new electronic software and systems based on the Internet. I hope I won’t surprise you when I say that hardly any of this has happened, that the developments that appeared most likely to emerge full blown in the early part of this year have evaporated into thin air. Perhaps it would be better to say that a few of these ideas have been put on hold to await what is now being called Act II of business-to-business.

The ideas whose remnants remain are those initiated by the financially strongest companies that are not quite yet willing to abandon such costly efforts, even though most have moved 180 degrees in their attitudes from believing that this new technology was going to change totally how wheat is traded and procured to wondering how this is finally going to happen. There’s a lot of hand-wringing about spending so much money on something that has so little immediate promise.

It’s probably correct to wonder whether any substantial commercial business in wheat is being conducted via the Internet and the systems that were designed to revolutionize the business of milling and of wheat trading. Of course, electronic trading is quickly coming to dominate trade in wheat futures, even as pit traders on the major U.S. exchanges raise hurdles to this changeover. But e-trading in futures, either wheat, agricultural commodities or financial derivatives, is not what the Internet promised flour milling (see related story, page 33).

On-line auctions never worked or gained broad acceptance; on-line exchanges did hardly better, and all the permutations in-between are now seen as just so much fluff. The Internet companies that had emerged with a business plan that seemed to promise great success have all but disappeared.

For the most part the glorious promise of the Internet as a medium that would revolutionize mill wheat procurement has all but vanished. Companies that should have known better dissipated huge sums on start-ups that hardly got by their initial multi-page press releases. These companies neglected the principle aptly expressed in an old Chinese proverb, "When you want to test the depth of a stream, don’t use both feet."

Like the technology and telecommunications manias, the concepts of Internet trading in wheat failed primarily because none either reduced costs or facilitated profits. No one successfully figured out how to make a return on what turned out to be massive investments costing far more than initial budgets, while revenues were hard to come by. Alas, the many promises proved illusory.

This should not be a surprise, considering the efficiencies that have been built into grain procurement over a long period of years. It turns out that there are few, if any, costs to be taken out of a system that already was so finely honed that even the most successful grain merchants or flour millers were unable to see savings in what the Internet offers.

Great profits in grain origination
and trading came from superior information that provided an advantage in
buying and selling. Knowing what
was happening before others in the
marketplace was the sure road to success.

But it wasn’t the Internet that wiped out this advantage. It was the gradual evolution of systems of communication that made information that once was guarded as highly private available to everyone. These systems were in place well before the Internet, but the global delivery of data facilitated by the Internet has so expanded this information availability that it’s the foolish trader who believes a piece of information is known by a few.

Let me now perhaps surprise you, though, by declaring that the Internet eventually will revolutionize the business in which you are now engaged. Despite all the gloom and doom pronounced about the worldwide web and the Internet, it is just too powerful a tool — combining communications with data manipulation at unbelievable speeds and with a limitless reach — not eventually to be a force in everything flour millers do.

I don’t limit that to wheat procurement. It’s going to influence how mills operate, how flour is distributed and how you relate to your customers, and they to their customers.

The Internet and the web are already a reality in ways that are commonplace in how business is done, such as e-mail, that we tend to overlook how very important it already is. Missing, though, is that ultimate revolution that quite literally will effect fundamental change in how your business is structured — how you buy wheat, mill it into flour, sell flour and the by-products, and deal with new products that the accompanying biotechnology will produce.

It will take time for all of this to come together, and if you want a prediction of how the Internet will unfold in wheat procurement, I would venture a guess that it will happen first in how farmers make decisions about what crops to grow, how to market them, and in either strengthening or broadening their market ties to millers and other buyers. The capabilities of the Internet offer so much to a universe like wheat farmers as well as to the numerous small bakers using flour that, in spite of all the fits and starts, it is here that Act II or maybe Act III will produce true revolutionary change.

As journalists and as historians of milling, we had hoped that the every-100-year cycles that have greatly affected milling of recent centuries would be reaffirmed by the Internet revolution.

It was at the start of the 19th century that milling first was automated by the introduction of bucket elevators and by the use of a central power source that made this industry the forerunner of the Industrial Revolution. And it was a century later that roller milling began in England and then spread like wild fire around the world to make for another revolution in the industry.

We’re still waiting for that next 100-year revolution. Without claiming total foresight, we very much think that the Internet will make this happen — not as quickly as might have been thought, but sometime in the not too distant future.


HARD REALITIES IN AMERICAN MILLING. Speaking of cyclicality of events in milling leads me to the hard reality that has hit American milling this year, and that is the reappearance of the excess capacity problem. In some ways, the excitement about the Internet’s possibilities diverted attention from what was happening in milling. In simplest terms, too much new capacity has been added in response to expectations of consumption growth that did not materialize.

In a capital-intensive industry like milling, that’s a bad combination. It has caused operating margins to deteriorate, plants to be closed and plant modernization projects to slow. Unfortunately, American millers as an industry find it difficult, if not impossible, to spend money on improving plants without also expanding capacity.

This is especially deadly when combined with new entrants into milling that elect not to buy existing plants, but to build new mills to serve markets that they have decided could best be served by establishing plants in locations where none existed before. Of course, this might have been less injurious than it has proved to be if consumption had continued to grow at the pace ruling when decisions were made about new plants. Not only did consumption stop growing, but it hit the proverbial stonewall.

Look at flour production, which reflects not just domestic consumption but also export business. Flour exports have become less and less important to American milling, accounting currently for 3% to 4% of annual production, but there are still mills that depend largely on foreign business. U.S. mills in 2000 produced more flour than ever before in history, slightly above 415 million hundredweights. That’s up 17% from a decade earlier and is 47% more than two decades ago, in 1980.

Using 1970 as the base year, we have traced the cumulative changes in annual production and annual capacity. For most of this period, production, driven by growth in domestic consumption, stayed well ahead of capacity increases, as millers exercised restraint. This restraint ruled until the late 1990s when, as a result of both new mill building and expansion of existing plants, the cumulative capacity increase moved ahead of production. Even this seemingly small change, with capacity expansion cumulatively only 1 million hundredweights ahead of production, undermined industry economics.

This is affirmed by the average rate of milling operations. We have observed that flour milling’s results are deemed satisfactory when mills run at 90% or better of six-day capacity. The annual rates of operations starting in 1980 show a decline in the last three years below the "magic" 90% rate.

The cumulative increase in capacity in the 1990s, at 73 million hundredweights, represents 44% of the capacity gains over the entire past three decades. Annual production in the 1990s rose only 61 million hundredweights, making for a clear picture of what went wrong.

Most of you are painfully aware that the reaction to this has been the shutdown of milling capacity. The three largest U.S. flour milling companies have closed eight mills with 78,000 hundredweights of daily capacity. Combined with the other four mill closings, a total of 102,500 hundredweights or 7.5% has been taken out of production.

I want to emphasize that these mills were shuttered, not as an unselfish contribution to the industry’s well-being, but as individual decisions aimed at improving the profitability of the companies. These companies seek to reduce their costs by closing their least profitable operations, in effect making their businesses stronger for competing in a fiercely competitive flour market that isn’t growing.

That the plant closings we’ve so far seen (which I guess is another way of saying we expect more) have mostly been done by the three largest companies prompts me to note how the excesses of the 1990s — if that’s the way to describe what has happened to American milling — reflected no indiscretion at the top.

The 1990s are notable for the largest companies actually seeing their share of total capacity shrink. From a peak of 23% in 1993, the largest milling company at the start of 2001 had hardly 20%, while the four largest companies have declined from a peak of 68% in 1993 to a low of hardly 61% at the beginning of this year.


MORE CHANGES UNDER WAY. Without predicting how 2001 will finally end, let me say that we believe the shares of the largest companies will fall even further. Such a trend is unique to milling in American industry at large, where the market shares of the largest companies have inexorably increased.

The message here is rather striking. It is that the largest companies in flour milling — which I need not remind you are divisions of corporate groups with a wide range of alternatives — have determined that milling is not a place to expand in the present environment. In some instances, investments are being made to build consumer brands in preference to processing businesses. In other cases, different parts of processing, such as oilseed crushing and corn refining, have won boardroom favor over flour milling.

And as most of you are aware, investment opportunities in flour milling outside of the United States have drawn the capital of several of the leaders in American milling, even leading the way to underscore that milling may be operated as a globe-circling business.

Changes are under way in the industries that are the major customers of U.S. millers, primarily the entire panoply of the business we call grain-based foods. Whereas the three largest flour milling enterprises are not the largest divisions of their parent corporations, bread baking in America has become the province of companies that bake bread as their principal enterprise.

This is a great shift from the period not so long ago when major bread bakers were, for the most part, subsidiaries of companies that were not even focused in food, such as brewing, telecommunications, liquor and other diverse entities. Now, for the first time in a long while, bread baking is the province of people who are experienced in baking.

Along this same line, though, it’s important to note that the two companies that account for more than half of the U.S. biscuit and cracker industry have recently come under new ownership. One can only be amused how Nabisco, once owned by RJR, and only recently acquired by Philip Morris as part of the latter’s commitment to growing its Kraft Foods business, will at last come out from under the tobacco ownership as Kraft separates from Philip Morris.

And then we have the pending merger of Pillsbury into General Mills, Inc., a change facilitated by a decision by U.K.-based Diageo that it prefers spirits to food, especially grain-based foods. When that merger is consummated, General Mills will be the world’s largest grain-based foods business from the viewpoint of consumer products, although its rank in U.S. flour milling is number six. We estimate that General Mills, with Pillsbury aboard, also becomes the world’s largest user of flour.

In reviewing how ownership changes have effected structural adjustments in the flour-consuming market, we are becoming increasingly aware of how transformation of the retail food business is affecting flour milling and even wheat growing. The emergence of Wal-Mart Stores, Inc. as one of our nation’s largest food retailers in the past decade stands as a development of huge importance.

Being able to respond to Wal-Mart’s demands when it comes to product and pricing has caused many changes in baking. One of these has to do with bakers needing to be able to assure Wal-Mart as to the quality controls in place among its suppliers, and I hardly need remind you that this increasingly has to do with genetically modified organisms, and how to identify their presence in the food chain.

Note that huge chains like Wal-Mart, as well as the other companies dominating the retail grocery business in America, have shifted the burden of inventory increasingly to food manufacturers. This is being facilitated in part by technology involving the Internet, aiming to integrate how food manufacturers deliver to central warehouses in coordination with the movement of product off store shelves.

In the case of perishable products like bread, especially with the dominance of direct store delivery by bakers, these pressures are less than for other products, but there’s no question that flour millers are finding a different set of demands from their baking customers as a result of changes at retail.

Practically every grocery store of any size in America has in-store baking, and for years in-store provided a spur to growth in flour consumption. Increases in this business have come to a halt in the past year or so, and this accounts in part for some of the pressures experienced
by millers.

Explanations for what has happened to in-store baking cover a wide range, but none seems more powerful than the absence of product innovation on the part of grocer-bakers. Indeed, all of baking is suffering from the absence of new products, in contrast to the tremendous boost in demand that flowed from the introduction and promotion of items like variety bread, muffins, bagels and artisan-
type loaves.


CONSUMPTION TRENDS. In discussing flour consumption trends, it’s important to understand that these measures are largely residual, or disappearance, numbers. We don’t have a good way of measuring consumption, which is sharply different from a usage number derived by measuring supply, made up of production plus imports, and deducting exports to arrive at disappearance.

Even this simple series has suffered in its accuracy from several complications. One is the disclosure that durum semolina output at mills integrated with pasta manufacturing plants, which accounts for a rising share of semolina output, had not been counted, because these so-called captive mills did not fit with the official definition of a flour mill. This omission was corrected at the start of 2000, but no changes were made for prior years, meaning that flour production in 1999 and before is understated by an unknown amount.

Let me raise a few more questions that are bothersome to us who believe that flour usage is an important benchmark for grain-based foods. The United States does a census of its population every ten years. The most recent, in 2000, has tentatively produced a number quite a bit larger (281 million vs. 275 million) than previous estimates.

If the larger number is accepted,
and there’s arguing about this among statisticians and politicians, it means an expanded denominator in calculating per capita disappearance of flour. The result would be lower per capita (145 lbs vs. 148 lbs) than previously estimated, not just for 2000, but also for most of the 1990s.

And then there’s the whole issue of how much flour calculated as disappearance is actually consumed. Heated debate is raging now over this issue, as statisticians try to translate per capita disappearance into actual per capita consumption. Estimates on wastage range between 30% and 40% of disappearance, which seems high.

The U.S. Department of Agriculture’s latest estimates place total flour disappearance at the record of 407.1 million hundredweights. That is up 5,045,000, or 1.3% from the previous year.

The problem is that durum semolina production alone increased more than the total gain in 2000, due to the inclusion for the first time of production by the integrated plants. We’ve tried to deal with this statistical aberration by estimating disappearance of flour minus durum semolina, saying in effect that we should not rely on the figures that understate semolina output.

Consumption, ex semolina, in 2000 actually decreased from the previous year, by a small amount. It was a near record, but not at the peak indicated by the aggregate numbers. Not so incidentally, this finding seems more in line with actual industry experience.

Per capita disappearance ex semolina in 2000 was at 136 pounds, down from 138 in the previous year and well below the recent peak of 139 pounds in 1997. If one wanted to stay with the raw government figures, per capita disappearance of all flour in 2000 reached 148 pounds, marking the second year in a row of one-pound increases. That 148 pounds would still be below the 50-year peak of 150 pounds attained in 1997.

All these numbers aside, though, it’s important to appreciate that the 1990s, even with the marked slowing in the last few years of the decade, still was marked by a significant increase in total and per capita disappearance of flour in the United States. That leaves us as the only developed nation to have experienced such growth.

Using annual averages for the entire decade, the rate of gain in the 1990s lagged slightly behind the soaring of the 1980s, but other than that represented a remarkable trend in U.S. flour usage.

The same goes for per capita disappearance, where the average of 144 pounds for all of the 1990s is nearly 20 pounds more than in the 1980s and is a good 30 pounds above the lows recorded in the 1960s and 1970s. This sharply differentiates the American experience from consumption trends in other developed nations.

The question this raises is, how did American grain-based foods bring about this enviable record? I will be forthright in saying that the industry had precious little to do with what happened. It came about largely as the result of two developments external to the industry.

One has to do with American lifestyles, which favor convenience and healthy eating. The other has to do with a sudden willingness on the part of the U.S. government to single out certain categories of food — grain-based foods, fruit and vegetables — over other foods in a change from decades-long governmental adherence to the "balanced diet" concept.

This is not the place for a review of how American living styles have been a great boon to grain-based foods. The desire for convenience is driven by many factors, none more important, though, than the large number of women who work. Awareness of the role of nutrition in health and well-being also ranks high. When combined with the government’s Dietary Recommendations for Americans and the Food Guide Pyramid, which emphasize that bread and other grain-based foods are the foundation of healthy eating, the industry had these many pluses in its favor.

Like the fisherman with a net in the middle of a stream full of fish, the industry couldn’t help but catch the benefit of developments that could only have been dreamed of in the not too distant past.

Quite a few people now believe these positives have either slowed or, at worst, expired as an influence in a nation where something new is constantly sought. The reference here is not just to a desire for new products, but also involves a country that constantly looks for new ideas about how lives should be lived, food prepared and eaten, and families tended to.

The vulnerability of grain-based foods to this slowing has been accentuated by the re-emergence of fad diets that promise quick weight reduction by eliminating complex carbohydrates. As a result of the widespread publicity about these diets, defining bread and flour as carbohydrates was transformed from being the positive mantra for the industry that it had been in the previous several decades to suddenly becoming a worrying negative. It’s easy to draw a direct line between the popularity of these quack diets and the slowdown in flour use.

All of this happening coincident with the over-ambitious expansion of milling capacity has created the present dismay, if not distress, that is rampant in our industry. With national milling operations averaging 88% of six-day capacity, returning to 91% requires a reduction not too different from the mill closings that have been already announced. The problem is that balancing capacity with demand, while a positive, is not enough to restore vitality to grain-based foods.

So far as we can tell, the Internet also will not restore the needed vitality. The failure of the Internet to assume an immediate major role in milling mainly reflects the industry’s long-standing efficiency — in procuring wheat, in milling flour and in distributing flour. As wheat origination has become integrated with milling and as the broker network that once reigned supreme in flour distribution has shrunk, the opportunities for the Internet to eliminate the middlemen have diminished.

There’s no question that the Internet will at some time assume a major role in flour milling, yes, in all of grain-based foods. How we’re not sure. We still believe the main opportunities lie in wheat procurement and in wheat segregation, which will become hugely important as genetic modification gains acceptance.

Yet, even with the wonders that the Internet will bring and with a newfound realism about balancing production and capacity, flour milling’s returns depend on a market that is growing, not just in line with population, but at a slightly greater rate. This is the goal toward which our industry must strive, looking for new ways to achieve this in collaboration with one another as well as in individual company efforts.


Review of 2000-01Flour Mill Closings

Daily flour milling

Date announced Company Mill location capacity (cwts)


ADM Milling Co.

Prtland, OR


January-Feb 2001

ADM Milling Co.

Buhler, KS


ADM Milling Co.

Inman, KS



Fisher Mills, Inc.

Portland, OR


Fisher Mills, Inc.

Modesto, CA



Con Agra Maple Leaf

Buffalo, NY



Cargill, Inc.

Springfield, IL


Cargill, Inc.

Topeka, KS


Cargill, Inc.

Jacksonville, FL



Heartland Wheat Growers

Russell, KS



Cenex Harvest States

Huron, OH



ConAgra, Inc.

North Kansas City, MO


Total capacity (12 mills)