This article was originally published by Murray Rothbard at The Mises Institute and adapted from “Bureaucracy and the Civil Service in the United States.”

Bureaucracy is necessarily hierarchical, first because of the Iron Law of Oligarchy, and secondly, because bureaucracy grows by adding more subordinate layers. Since, lacking a market, there is no genuine test of “merit” in government’s service to consumers, in a rule-bound bureaucracy seniority is often blithely adopted as a proxy for merit. Increasing seniority, then, leads to promotion to higher ranks, while expanding budgets take the form of multiplying the levels of ranks under you and expanding your income and power. Bureaucratic growth occurs, then, by multiplying levels of bureaucracy.

The theory of hierarchical government bureaucracy is that information is collected in the lowest ranks of the organization and that at each successive higher rank, the manager culls the most important information from his subordinates, separates the wheat from the chaff, and passes the culled information higher up, so that, in the end, the President, for example, dealing with intelligence operations, receives a two-page memo distilling the most important information gathered and culled from hundreds of thousands of intelligence agents. The President, then, knows more than anyone else, say, about foreign affairs. One problem with this rosy model, as Professor Gordon Tullock points out in his illuminating book, The Politics of Bureaucracy, is that the model doesn’t ask whether or not each bureaucrat has the incentive to pass the best distillate of truth on to his superiors. The problem is that bureaucratic favor, especially at the higher levels, depends on pleasing one’s superiors, and pleasing them largely rests on telling the President and the higher bureaucrats what they want to hear. One of the great truths of human history is that one tends to shoot, or at least react badly, to the bearer of bad news. “Sire, your policy is working badly in Croatia” is not the sort of message that the President, say, wants to hear from his envoy, and while the outcome in Croatia remains in doubt, the President and his aides want to continue to believe that their policy is doing well. Hence, the dissident is set down as a trouble-maker if not a subversive, and his career in the hierarchy is side-tracked, often permanently. In the meanwhile, the envoys or foreign service people who assure the President “things are going very well in Croatia” are hailed as perceptive fellows, and their careers are advanced. And then, if years later, the dissident is proved correct, and the Croatian policy lies in shambles, is the president or any other ruler likely to turn in warm gratitude to the former dissident? Not hardly. Instead, he will still remember the dissident as a troublemaker, and he will not blame his aides, who, along with himself, have been proved wrong. For, after all, didn’t the great mainstream of experts make the same error? How common is sincere soul-searching and repentance for past errors among Presidents or other rulers?

Those bureaucrats who are shrewd analysts of human nature, then, and who understand the way rulers operate, will, if they see that the cherished policy of their President is in grave error, tend to keep their mouths shut and let some other sucker be the messenger of bad news and get shot down.

Every human activity and institution will tend to reward those who are most able to adapt to the best route to success in that activity. Successful market entrepreneurs will be those who can best anticipate and satisfy consumer demands. Success in the bureaucracy, on the contrary, will go to those who are most apt at (a) employing propaganda to persuade their superiors, the legislators, or the public about their great merits and, therefore, (b) at understanding that the way to rise is to tell the President and the top bureaucrats what they want to hear. Hence, the higher the ranks of the bureaucracy, the more yes-men and time-servers there will tend to be. The President will often know less about what is going on than those in the lower ranks.

Hence, for example, the phenomenon of President Nixon, thinking he knew more than anyone else about the Vietnam War and yet actually knowing less than the astute reader of the New York Times. For the CIA and other intelligence warnings of what was going on, developed by many of the lower officers, were screened out by the higher-ups, for being contrary to the President’s preferred line, i.e., that all was going well.

The standard explanation of why government grows is that, as time goes on, there is more work for government to do, and that therefore the public’s “demand for government” rises. Far more accurate is the view that there is a case of an inverted Say’s Law, where supply — or rather the suppliers of government “services,” the bureaucracy — themselves constitute the “demand” for their own services, and that they engineer the consent of their superiors, or of the legislature, to provide the wherewithal in the form of increased taxation. Contrast the hilariously satirical, but all too perceptive account of “Parkinson’s Law” of bureaucracy. Thus, Professor Parkinson asserted that, in a government bureaucracy, “there need be little or no relationship between the work to be done and the size of the staff to which it may be assigned.” The continuing rise in the total of government employees “would be much the same whether the volume of the work were to increase, diminish, or even disappear.” Parkinson identifies two “axiomatic” underlying forces responsible for this growth: (1) “An official wants to multiply subordinates, not rivals”; and (2) “Officials make work for each other.”

Parkinson begins his “model” with an official who feels himself overworked. The official could resign, but that is unthinkable; besides, he would lose his pension rights. To ask to divide his work in half with a new colleague on his own level is equally unthinkable; for his status would be cut, and he would bring in a dangerous rival for the job of his own boss when the latter retires. He could ask for one assistant under him; but that would be dangerous, because the new man might achieve something like equal status with himself. No, his preferred route is to ask for two assistants, who could then compete with each other for his favor; pretty soon, each of these new assistants will complain of overwork, and each one of these will get two assistants. The original bureaucrat now has the satisfaction of having six men under him, and he is now ready for a promotion and a substantial raise in pay.

But how about the work to be done? Won’t the original quantity of work be divided into seven parts, and won’t each man now be absurdly and manifestly idle and underworked? No — and here is one of Parkinson’s scintillating insights into the theory of bureaucracy — for one aspect of Parkinson’s Law is that “work expands so as to fill the time available for its completion.” Or, as Parkinson also puts it: “The thing to be done swells in importance and complexity in direct ratio with the time to be spent.” Here enters the second aspect of Parkinson’s Law of Growth: that “officials make work for each other.” For, says Parkinson, “these seven make so much work for each other that all are fully occupied,” and the original man “is actually working harder than ever.” Documents have to be sent to each man in turn, each has to comment on the document and send the comments to everyone else, they all have to confer on the document and the various amendments proposed, and the original man is now also wrapped up in problems of inter-personal relations between himself and his staff, and of each of his staff amongst the others. Finally, after a lengthy process of interaction, writes Parkinson, the original official produces the same reply to the document that he would have written if all his subordinates “had never been born.” “Far more people,” Parkinson concludes, “have taken far longer to produce the same result. No one has been idle. All have done their best.”

Parkinson then illustrates his law with delightful examples from the British Royal Navy. From 1914 to 1928, the number of ships in the Navy fell by 68 percent; the number of officers and men fell by 32 percent. And yet, during the same period, the number of dockyard officials and clerks in the Navy increased by 40 percent, while, even more outrageously, the number of Admiralty officials increased by over 78 percent. The annual rate of increase in the number of Admiralty officials, with little variation, was 5.6 percent. Parkinson takes another example from the British Colonial Office, from 1935 to 1954. In that period, the area and population of colonial territories remained about the same from 1935 to 1939, fell during the war until 1943, rose again until 1947, and then steadily decreased as Britain shed its Empire. And yet, in each of these two decades, the Colonial Office bureaucracy rose steadily in number by about 5.9 percent per year, regardless of what was happening in the scope of the alleged work to be done. Considering then the rate of increase each year in the Admiralty, and averaging the rates of increase of Admiralty and colonial officials, which is not, after all, more outlandish than many other statistical procedures, Parkinson triumphantly concludes that the number of officials will increase by an average of 5.75 percent per year, “irrespective of any variation in the amount of work (if any) to be done.”

A similar analysis was set forth earlier, in 1950, in a grievously neglected book by Connecticut attorney and farmer Thomas H. Barber, based on years of inquiry into government and on his observations of Washington bureaucracy during World War II. Barber writes that “there are two requisites for a bureaucrat’s promotion, the first, the ability to get and hold votes, the second, the number of subordinates he is able to keep busy.” Barber goes on:

… in the Federal Government the pay of a bureaucrat executive is proportioned by Civil Service law to the number of his subordinates. This leads to the rivalry in Washington as each bureaucratic chief tries to increase his “empire.” Generally, in order to keep his subordinates busy the boss assumes an air of great importance and affects to be very hurried and under great pressure. He is very punctual at the office and insists that everyone else be. He then deliberately begins to multiply paperwork, calling for reports on any subject connected with his job. He issues enormously complicated orders and memoranda for the organization of his office, requiring that all papers be so routed round that almost every scrap has to be read by everyone in the office and discussed by a number of interlocking committees before it is acted upon. He requires that no paper be thrown away, but all shall be cross-indexed and filed. He has anybody who can be tagged, interviewed, a stenographic report made of the interview and typed (often he has them mimeographed), and circulated to be read and initialed. By these methods it is quite easy to take an amount of work that could be done easily and efficiently by three men and two stenographers, and blow it up so that it can keep from fifty to two hundred people extremely busy, and yet fall far behind in its execution. Thus the uncompleted work gives him an apparently sound excuse for more clerks, who increase his prestige and his pay.

Barber then goes on to relate a delightful example of bureaucracy in action that he had observed during World War II. He notes that there existed a department whose work, “supposing it was worth doing, which is doubtful,” could have been done competently by about twenty people. It was run, as he puts it, “by a man with a bureaucratic soul.” This man asked for written opinions from everyone on all sorts of subjects and had everyone read and initial them:

He was always intensely busy himself, even at night; and he kept constantly increasing his department till he got it up to two hundred men and women. This made him very important. All the two hundred were so busy carrying out his regulations that they were in a constant sweat and confusion, had no time to think, and the essential work in support of the war effort — supposing it was essential — suffered dreadfully. He was rewarded and translated to a more important job.

His successor, Barber related, was a different kind of person; an old gentleman with little ambition and little regard for the taxpayer but whose objective was to do the essential work and keep himself and everyone else at the workplace contented. In contrast to the twelve hours a day spent in the office by his predecessor, this man spent only one-half an hour at work each morning. The rest of the day, he walked around the office, talking and joking with the employees, and played golf in the late afternoon. At the end of the first week, says Barber, “he fired about fifty of the two hundred people, apparently at random.” As a result, “the work lessened considerably for the remaining ones.” There was naturally a lot of discussion of this action, and “it was generally decided that he had fired the fifty he was sure he did not like.” “Not a very scientific way of eliminating surplus help,” adds Barber, “but it did lighten the work.”

The following week, the new boss fired fifty more people, this time apparently dismissing those “he thought he did not like.” In consequence, “the work for the remainder lightened enormously, though some of the essential work of those dismissed was apportioned around silently by those that remained.” A few days later, another fifty people were dismissed, these being the people “he was not sure he did like.” Barber notes: “With three-quarters of the force eliminated there was practically nothing left but the ‘essential work’, such as it was, to do.” This work was done effectively in about half of each day by the fifty people remaining, “far more efficiently than it had been done by the original two hundred. The fifty did their stuff and devoted the time remaining — about half of it — to their own concerns.”

Barber concludes that “the old gentleman, being now surrounded only by those he knew he liked, felt he had done enough.” He was in the office about an hour a day, and then he evaporated. “The ‘work’ was far better done than it had been, people had time to think and were not in each other’s way.” Barber adds that the work probably could have been done by half again of those remaining, but that then the half would “have had to work about as hard as the original two hundred had worked and there would have been no benefit to anyone but the taxpayers.”

In addition to this keen treatment of bureaucracy, Thomas Barber was perhaps the first person to arrive at the essence of what is now called “public choice” analysis in the economics profession. Barber notes the “constant tendency for all governments to grow both in size and in authority.” Why? Barber answers:

because the advantage of a big, powerful government, from the point of view of the bureaucrats, is personal, clear and ever-present to their eyes; and because the cost of it, not only in money but in freedom, which is lost by giving authority to officials, is vague and nebulous in the minds of the citizens whose attention is not focused on the government at all…. Therefore, since the bureaucrats know exactly what they want and are working for their own immediate interest, and since the other citizens do not realize what they are giving up, and, in fact, have not their attention on the matter at all, it is obvious which group will prevail.

What public choicer has put it better?