Jane Jacobs, The Economy of Cities.
Random House Trade; (February 1, 1970)
Today, only two cities in all of Britain remain economically vigorous and prosperous. One is London. The second is Birmingham. The others have stagnated one by one, much as Manchester did, like so many lights going out. British town planners, ironically, have regarded London and Birmingham as problems, because they are places in which much new work is added to old and thus cities that persist in growing. The British New Towns policy was specifically devised to discourage the growth of London and Birmingham and "drain it off." Birmingham's economy has remained alive and has kept up to date. Manchester's has not. Was Manchester, then, really efficient? It was indeed efficient and Birmingham was not. Manchester had acquired the efficiency of a company town. Birmingham had retained something different: a high rate of development work.
Efficiency as it is commonly defined -and I do not propose to change its definition, which is clear and useful -is the ratio of work accomplished to energy supplied. We can speak of high or low rates of efficiency because, in any given instance, we have two relevant factors to measure: input of energy, and quantity and quality (value) of work accomplished. We can compare the measurements in one instance with measurements in other instances. Manchester turned out a great deal of cloth relative to the energy supplied by its workers and by those who served the needs of the workers in the city.
But these particular measurements are not relevant when development work is wanted. A candy manufacturer, reminiscing to a New Yorker reporter about the first candy bar he developed as a shipping clerk in a candy factory, recalls, "I showed it to my boss and he was very happy. `How many of these can you make in a minute?' he asked me. `In a minute?' I said. `It took me four months to make this one!' " Suppose it had taken him eight months? Or two months? That measurement has nothing to do with the operating efficiency envisioned by his boss.
Efficiency of operation, in any given case, is a sequel to earlier development work. Development work is a messy, time- and energy-consuming business of trial, error and failure. The only certainties in it are trial and error. Success is not a certainty. And even when the result is successful, it is often a surprise, not what was actually being sought.
A low rate of efficiency in production work means that the person or organization doing the work is going about it ineptly. But the exorbitant amounts of energy and time and the high rates of failure in the process of developing new work do not mean the development work is being done ineptly. The inefficiency is built into the aim itself; it is inescapable. There is no systematic way to evade it. The president of DuPont, a company that has tried to systematize its development work to the highest degree possible, has told a Fortune reporter that only about one out of twenty of those research projects that the company decides to develop further after initial exploratory work turns out to be useful to the company. The fact that an organization engages in large-scale production, which is what makes a large organization large, and that it produces very efficiently too, does not mean that the efficiency spills over into development work.
Indeed, development work is inherently so chancy that by the law of averages, chances of success are greatly improved if there is much duplication of effort. The U.S. Air Force's analytical organization, the Rand Corporation, having been assigned to study how waste could be eliminated in the processes of military development work, came to the conclusion that although duplication of effort was theoretically wasteful, it was not wasteful empirically. For one thing, the report said, different people brought different preconceptions to development work and there was no way of telling in advance which might prove fruitful or where it might lead. Eminence or reputation or even past success was not a reliable indicator. The report cited, as an illustration, the fact that in 1937 when the jet airplane engine had already been developed in Britain (largely in Birmingham, as it happens), a committee of distinguished aeronautical experts in the United States, to whom this event was not yet known, having studied the possibilities of jet propulsion, came to the conclusion that it was not practicable. It was their recommendation that attempts to develop jet propulsion he dropped. The Rand researchers said that they had found definite waste, and a lot of it, in the development work of the military establishments; it was the great waste of administrative man-hours and energy devoted to trying to eliminate duplicated effort. Just so, when Pasteur, that wise old man, begged for enlarged support of the biological sciences, he begged for multiplication of laboratories.
Is it not possible for the economy of a city to be highly efficient, and for the city also to excel at the development of new goods and services? No, it seems not. The conditions that promote development and the conditions that promote efficient production and distribution of already existing goods and services are not only different, in most ways they are diametrically opposed. Let us consider a few of them.
Breakaways of workers-especially very able workers from existing organizations promote the development of new work as well as the creation of new organizations. But breakaways, are not good for the parent company; they undermine its efficiency. To the company or companies in control, one of the advantages of a company town is that breakaways are not feasible there. And in any settlement where breakaways are inhibited, by whatever means, the development rate must drop, although the efficiency of already well-established work is apt to climb.
Now consider for a moment the question of suppliers of bits and pieces of work to other producers. Many relatively small suppliers, much of whose work duplicates and overlaps, are indispensable to a high rate of development. But they are not efficient, neither in respect to their own work nor the operations of the producers who buy from them.
Consider also the conflict between development and efficiency as it applies to the work of investing development capital and supplying working capital. The most efficient way to invest capital (whether by government, by semipublic, or by private lenders and investors-it does not matter) is through a relatively few large investments and loans, not through many small ones. If small loans are made, it is most efficient to consolidate them, in effect, by making them only for purposes that have already become standardized and routinized. To put capital into the purchasing of enterprises that produce goods and services already developed is more efficient than to put it into development of new enterprises and new work. Also, it is efficient to invest development capital in a sure thing-if in new work, then in new work for which customers are guaranteed in advance.
It is most efficient for large construction firms to produce monotonous multiples of identical buildings; it is most efficient for architects to design multiples of identical buildings. Superblocks are more efficient than smaller blocks because there are fewer crossings and traffic can flow more efficiently; when there are fewer streets, utilities can be distributed more efficiently and of course the maintenance of streets costs less. Indeed, numerous small enterprises, just by existing, are in conflict with the economic efficiency of a city's large and well-established enterprises.
Earlier in this century, it was conventionally supposed by American philanthropists that poverty is caused by disease. Healthy people, it was reasoned, would be more productive, have more initiative, be more capable of helping themselves, than people in ill health. Poverty was analyzed as a vicious circle in which poverty leads to disease and disease reinforces poverty. Measures to combat disease turned out to be quite successful at combating disease, irrelevant for combating poverty. They helped lead to the situation that is now being diagnosed as a different vicious circle-poverty-overpopulation-poverty. To seek "causes" of poverty in this way is to enter an intellectual dead end because poverty has no causes. Only prosperity has causes. Analogically, heat is a result of active processes; it has causes. But cold is not the result of any processes; it is only the absence of heat. Just so, the great cold of poverty and economic stagnation is merely the absence of economic development. It can be overcome only if the relevant economic processes are in motion. These processes are all rooted, if I am correct, in the development work that goes on in impractical cities where one kind of work leads inefficiently to another.