Death and Taxes: The Origins of Federal Nosiness


In a message to Congress in June 1934, President Franklin Delano Roosevelt outlined the goals and accomplishments of his administration in dealing with the nation's Great Depression. "Among our objectives," Roosevelt said, "I place the security of the men, women, and children of the Nation first. This security for the individual and for the family concerns itself primarily with three factors. People want decent homes to live in; they want to locate them where they can engage in productive work; and they want some safeguard against misfortunes which cannot be wholly eliminated in this man-made world of ours."

The "safeguard against misfortunes" that Roosevelt envisioned was a national system of social insurance, funded by employee contributions and administered by the federal government to ensure a stable and long-lasting source of funds for old age. On August 14, 1935, Roosevelt achieved his goal by signing into law the Social Security Act (SSA).

A main goal of the SSA was to address one of the myriad problems facing the country during the Depression—the steadily declining economic status of its elderly. For centuries, when people reached retirement age or got sick, they were cared for by their extended families, who usually lived within shouting distance. With the growth of social mobility (spurred by the intercontinental railroad), families began to split and move around the country. One consequence of that trend was that the Great Depression hit the older segments of the population particularly hard. In large part, the SSA was the federal government's effort to provide a basic social safety net for its more vulnerable older citizens.

From the moment that the Social Security Act was first proposed, the concept of social insurance and the mechanisms chosen to implement it became the subject of extensive debate. Most of those debates are outside the scope of this book, but two aspects of the Social Security Act are directly relevant to the issue of workplace privacy: the tracking of employee earnings to calculate benefits and the creation of a system of Social Security numbers (SSNs). In drafting employers to help the government collect not merely information about employees but also their social security contributions, Congress got a seductive taste of how useful businesses could be in the implementation of policy initiatives.

More importantly, the creation of the Social Security number unwittingly ushered in a new era of information tracking by the federal government. For the first time, Washington had a means—albeit flawed—to link individuals to a potentially limitless number of records. In the precomputer days of the late 1930s and early 1940s, there was no way to anticipate the extent to which the SSN would infiltrate American society, any more than there was a way to predict that a post-World War II baby boom would threaten the continued viability of the Social Security trust fund itself. Nevertheless, there is little question that the Social Security number has become our de facto national ID; along the way, it has become a powerful tool for facilitating other Congressional invasions of workplace privacy.

Collecting Data for the Social Safety Net

Appropriately enough, the Social Security Act is itself a senior citizen—it first took effect sixty-five years ago. On January 1, 1937, employees began accruing credits toward old age insurance benefits under the terms of the Social Security Act.

In order for employees to receive payments under the SSA, it was necessary for the federal government to collect detailed information about their earnings. While this was not the first time the federal government had collected information about employee earnings (the government briefly instituted a system of payroll withholding in the nineteenth century), the system for collecting employee data under the Social Security Act has proven particularly durable.

Setting up the new program, in an era before computers, was a phenomenal logistical challenge. With the January 1, 1937, deadline looming, the newly created Social Security Board essentially hired the United States Post Office to help it distribute and collect the necessary paperwork. First, the Board mailed enrollment forms to the nation's 3.5 million employers; when the forms were returned, the Board assigned each employer a unique Employer Identification number (EIN). At the same time, application forms for a Social Security number were being sent out to employees around the country.

When employees returned their applications, their local post offices assigned them a number from a pool allocated by the Social Security Board to each post office. The completed application forms were then handed over to the nearest local Social Security field office, and then forwarded to the main record-keeping facility in Baltimore, Maryland. By June 1937, the Social Security Board had set up 151 field offices around the country and had assigned more than 30 million SSNs.

On the Social Security Administration website, you can view an image of the original SS-5, the form used by employees to apply for a Social Security number. The form required each employee to provide a variety of information:

  • First, middle, and last name (Married women were instructed to "give maiden first name, maiden last name, and husband's last name.")

  • Address

  • Current employer's name and address

  • Age, date of birth, and place of birth

  • Sex and color ("white," "negro," or "other")

  • Registration number, if any, with the U.S. Employment Service [1]

  • The location and date where the applicant previously filled out an SS-5

Prior to the arrival of the SS-5, the only employees in the United States with direct experience in answering questions posed by agents of the federal government were those inducted into the military or suspected of a federal crime. But in just six short months, the U.S. government successfully collected information from 30 million people about their names, their race, where they lived, and for whom they worked.

At that time, employees had a choice: They could decide whether answering the questions was worth the prospect of receiving benefits after they retired. Today, as a practical matter, that choice no longer exists. When the Internal Revenue Service began using the Social Security number in 1961 as the taxpayer ID number, the SSN became a required element of any financial transaction, including employment. In theory, providing your Social Security number is a "voluntary" act, but in reality, the chances of your being hired without a Social Security number are fairly small. A typical view is expressed in the privacy policy on the USAJOBS website, a online resource created and maintained by the U.S. Office of Personnel Management:

Giving us your SSN or any of the other information is voluntary. However, we cannot process your application, which is the first step toward getting a job, if you do not give us the information we request. [2]

The passage of the Social Security Act was a momentous event for our society as a whole, but it was also a pivotal moment in the relationship between the nation's employers and Washington, one that helped usher in an unprecedented era of data collection by federal and state governments.

Executive Order 9397: The Birth of a National ID

The Social Security Act was not greeted with universal approval; there was widespread concern about the possibility that the Social Security number would become a national identifier and the uses to which it might be put. During the 1936 election, media mogul William Randolph Hearst published a story stating that Americans would be issued a steel "dog tag" with their names and Social Security numbers, and that they would be required to answer highly personal questions on the Social Security applications.

Hearst and his newspapers had a well-deserved reputation for outright fabrication when it suited the bombastic publisher's purpose, but in this case, there was a small kernel of truth. The idea of metal Social Security IDs had in fact been presented to the Social Security Board by the Addressograph Corporation, which was keenly interested in bidding on a contract to produce millions of metal tags. Although the Board categorically rejected the idea, Addressograph made up a sample tag for SSA Board member Arthur J. Altmeyer, who kept the tag as a souvenir, eventually donating it to the SSA's History Room.

Hearst may have been a semiparanoid yellow journalist, but his concern over the SSN ultimately proved correct. Just six years after the implementation of the Social Security Act, President Roosevelt issued Executive Order 9397, which established the Social Security number as the official method of identifying individuals by federal agencies and departments. Any federal agency needing to identify people was ordered to use their Social Security numbers, and the Social Security Board in turn was ordered to provide any Social Security numbers an agency requested.

As a result of Roosevelt's Executive Order, the SSN is now an essential element of our interactions with both federal and state governmental agencies. At least forty different federal agencies or programs require use of the Social Security number for individual identification, including the Internal Revenue Service (which began using the SSN in 1962), the Department of Defense (1967), Aid for Families with Dependent Children (1975), the Selective Service System (1981), and the National Student Loan Data System (1989).

Of the various federal agencies that have adopted the Social Security number as a sorting tool, it is really the IRS that has done the most to solidify the SSN as our national identifier. Under IRS regulations, everyone over the age of eighteen who earns income is required to have a Social Security number, and employers are required to use that Social Security number to report the payment of income. More recently, the IRS adopted regulations requiring parents to obtain Social Security numbers for children they claim as dependents on their tax returns.

Congress has made some half-hearted efforts to limit the use of the Social Security number as a national ID. In 1974, for instance, the reformminded congressmen who were elected following the Watergate scandal adopted the Privacy Act (5 U.S.C. 552a et seq.), which purported to regulate the handling of personal data by federal and state agencies, declaring that "... in order to protect the privacy of individuals identified in information systems maintained by Federal agencies, it is necessary and proper for the Congress to regulate the collection, maintenance, use, and dissemination of information by such agencies."

Despite Congress's professed concern over the widespread use of the Social Security number as an identifier, however, legislators have continued to require or permit its use. In addition to the programs mentioned above, half of which post-date the Privacy Act, Congress also adopted legislation in 1976 that specifically permitted states to use Social Security numbers as identifiers on motor vehicle registrations, tax reporting forms, and drivers' licenses.

In 1999, Representative Ron Paul (Texas) proposed the Freedom and Privacy Restoration Act, a bill that would bar the creation of any "uniform standard of identification," prohibit any federal agencies from using the same number to identify a particular individual, and bar the use of the Social Security number as an identifier.

The Subcommittee on Government, Information, and Technology of the House Government Reform Committee held hearings on Representative Paul's bill, but took no further action. Paul reintroduced his bill last year as the "Identity Theft Protection Act of 2001" (H.R. 220), and the proposed legislation was referred to the Subcommittee on Government Efficiency, Financial Management and Intergovernmental Relations, where it languished for the remainder of the 107th Congress.

Despite Representative Paul's efforts, the fact of the matter is that the Social Security number has effectively become a national ID, an outcome that is underscored by comparing today's SS-5 with the original. Unlike the 1936 edition, the current Social Security number application form makes no reference to employment at all, implying that its use is no longer limited to the original purpose of identifying an employee's contributions account to the Social Security Administration.

As an identifier, the Social Security number has some serious flaws:

  • It's too short—although nine digits theoretically offer 999,999,999 different combinations, the Social Security Administration may run out of unissued numbers sometime around the end of this century.

  • It's insecure—with roughly 450 million Social Security numbers having been issued since 1935, the odds are roughly one in two that any randomly generated nine-digit number is a valid Social Security number (and those odds can be improved by looking at the SSA's helpful table of the areas to which the first three digits have been assigned).

  • It's inconsistently applied—the Social Security Administration does not automatically issue a Social Security number to everyone who is born in or enters the U.S., waiting instead for people to request a number.

  • It's not self-verifying—unlike more well-designed numbering systems, the Social Security number lacks a check digit, a number appended to the end of the string of numbers that verifies that the substantive numbers in the string are in fact correct. [3]

Efforts to create a consistent, well-thought-out national identifier were loudly shouted down in the late 1940s and 1950s. Within a decade, however, private businesses were computerizing their records and looking for some tool to help them identify individuals in their electronic files. Despite its obvious shortcomings, the SSN gained widespread use as an individual identifier because there was no other viable alternative when computers entered the workplace. A unique number is far more effective in distinguishing individuals than names (which have an annoying tendency to repeat, be misspelled, change through marriage or adoption, and so forth).

The consequences of our unplanned adoption of the Social Security number as a national identifier have been profound. Not only does the SSN serve as a handy tool for uncovering and tracking large amounts of information about nearly everyone in the United States, employees and nonemployees alike, but it has also given rise to the Digital Age phenomenon of identity theft.

If someone can learn the name and Social Security number of another individual, they can pretend to be that person for the purpose of obtaining credit cards and other financial resources. Identity theft has become a vast problem for the credit industry, banks, and individuals; in 2000 alone, Mastercard and Visa reported identity theft losses of $114 million, a 43 percent increase from the year before. Law enforcement, using a broader definition of identity theft, estimates that the payment card associations actually lost a total of $1 billion to identity theft in 2000. [4] The impact on the victims ranges from the annoying—the tedium of replacing credit cards, licenses, etc.—to the devastating—years of contesting huge bills and rebuilding personal credit.

The credit card companies' response to the increasingly vast problem of identity theft has been, to put it charitably, half-vast. Far too many card offers are sent willy-nilly through the mails, seeking only a name and Social Security number in exchange for a freshly printed credit card. Some preliminary steps have been taken to increase the level of identification required to issue a card (which in itself raises privacy issues), but it still remains disturbingly easy to obtain credit cards under false pretenses.

Federal Income Tax Withholding

Next to our medical history, how much money we make (or don't make) is probably the most sensitive topic for most Americans. For the last ninety years, however, anyone who earns money in the United States has been required to report to the government how much she made in the preceding year and to pay the taxes that are due on those earnings.

For the first thirty years of the modern income tax (which began when the Sixteenth Amendment—authorizing an income tax—was passed in 1913), the government relied largely on the honor system. At the end of each year, Americans were required to calculate their taxes on the preceding year's income and make a lump sum payment of the amount due.

As the enormous cost of fighting World War II began to have an impact on the U.S. Treasury, a movement arose to withhold estimated tax payments from wages. This was not the first time that withholding had been used to collect taxes: In 1862, the cost of the Civil War led to the adoption of the first income tax in the nation's history, and Congress included a withholding provision to capture tax revenue at the source of wage payments. Congress repealed that tax in 1872. A withholding provision was also included in legislation that followed the adoption of the Sixteenth Amendment, but that provision was dropped in 1917 after it led to extensive criticism of the new taxation system. It was not until 1943, when Congress passed the Current Tax Payment Act, that the system of withholding that we know today first went into effect.

Unlike earlier attempts to impose withholding, the resistance to the Current Tax Payment Act was far more muted. The country was distracted by World War II, and advocates of withholding argued strongly that it would increase the collection of revenues. There's no question that it did so: By 1945, after just two years of tax withholding, collections had risen from $7 billion to $43 billion.

The biggest reason for the successful introduction of withholding, however, was that the essential framework for a withholding system was already in place. Thanks to the Social Security Act, the federal government had seven years' experience in operating a payroll tax system (even though the Social Security payments were called a "contribution" instead of a tax). In addition, in 1942, the U.S. government had instituted a "Victory Tax," a 5 percent surcharge on everything earned above $624. Employers were required to withhold the tax and make periodic payments to Washington.

Sixty years after its inception, tax withholding is a firmly established part of our governmental system. The fact that the government effectively gets an interest-free loan from each worker is the subject of numerous impassioned articles in conservative publications. Remarkably, however, little attention is paid to the fact that tax withholding, along with the Social Security system, has created an institutional machinery of inquisitiveness by the nation's employers (which now number more than 6 million). In order to comply with federal record-keeping and withholding requirements (at an estimated annual cost of $5,100 per employee), employers are required to collect and maintain extensive records for each employee. More importantly, employers are required to share those records with a wide variety of governmental agencies, few of which have a good track record for preserving private information. [5]

Without one significant invention, the impact of governmental reporting requirements and poor security on employee privacy would be minimal. But the digital computer has made it possible (particularly in conjunction with the Social Security number) to compile incredibly detailed profiles of individuals both in and out of the workplace. While a better-informed government is arguably one that can better plan and implement policy, it's reasonable to ask whether greater governmental efficiency is an acceptable trade-off for markedly less workplace privacy.

Data Collection

In 1890, the United States government was facing its first major data-processing crisis. Under the terms of the United States Constitution, the government is required to conduct a census of its citizens every ten years. The 1880 census, tabulated entirely by hand, had taken seven and a half years to complete, and census officials seriously questioned whether it would be possible to complete the tabulation of the 1890 census before the start of the next census in 1900.

Looming on the horizon, however, was a technological solution that would take the census into the new century, and usher in an era of unprecedented data collection and manipulation. As the tabulation of the 1880 census began, a young man named Herman Hollerith went to work for the Census Bureau. Hollerith was familiar with the use of punch cards by Frenchman Louis Jacquard to produce complex woven patterns in his mechanical looms, and Hollerith thought that a similar process could be used to speed up the tabulation of the census. Over the course of the next decade, Hollerith perfected his design and when his machine won a tabulation contest in 1890, Hollerith was awarded the contract to count the 1890 census.

In an era of microchips and flat-panel plasma displays, the design and operation of Hollerith's machine seems positively quaint. Census data collectors used traditional paper forms to collect information. Those forms were sent back to Washington, where the information on the forms was put onto dollar bill-sized cards, each of which had twelve rows and twenty columns of holes that could be punched out to represent the collected data. The cards were then fed into Hollerith's tabulating machine, which was equipped with 240 metal pins, one for each of the possible holes in the data card. When a pin lined up with a hole in the punch card, it would make contact with a tiny cup of mercury below the card, completing an electrical circuit, and increasing a corresponding dial on the machine by one notch.

As archaic as it seems today, Hollerith's tabulating machine represented a phenomenal advance in counting technology. Thanks to his invention, the 1890 census was completed in just two and a half years—despite a 13 percent increase in the U.S. population. [6]

The data punch card became a mainstay of government and private business for the better part of a century. Beginning in 1906, for instance, railroads across the country began using the punch card to maintain their operational records and run their businesses more efficiently. Four years later, Aetna Life and Casualty became the first insurance company to use punch cards and Hollerith machines to compile mortality statistics.

The success of the Hollerith machine in processing the census records for 65 million people made it abundantly clear to other federal agencies that punch cards could be used in a variety of different ways. During World War I, for instance, the U.S. Army used punch card technology to keep track of supplies, equipment, and personnel records (including both medical and psychological records). By 1933, punch card technology had become so pervasive in federal government operations that they were being used by Washington as checks—the Agricultural Adjustment Administration issued the first punch card check that year, and from its start in 1937, Social Security Administration checks were also punch cards. [7]

In this modern age of data storage, when the laptop computer on which I am writing this book can store approximately 37,000 copies of the finished manuscript, it's difficult to appreciate the sheer physical mass of the punch cards needed to hold records created by the implementation of the Social Security Act alone. According to a research note prepared by the Social Security Administration's historian, in 1937 it took 24,000 square feet of storage space to hold all of the necessary employee and employer punch cards and related paperwork. It turned out that there was no building in Washington with floors strong enough to hold the nation's Social Security records, so they were stored for a time in an old Coca-Cola factory in Baltimore.

Snow White to the Rescue [8]

The physical mass of records stored on punch cards was only the beginning of the logistical problems. In the months and weeks leading up to the implementation of the Social Security system, there was widespread skepticism that the Social Security Board could successfully assign all of the necessary numbers and do anything meaningful with the huge amounts of information that it would be collecting. According to the Social Security Administration's own history, a French industrial expert who had been hired to advise the U.S. on the implementation of the Social Security Act flatly declared that the program could not be implemented.

The solution, appropriately enough, was provided by an economic descendent of Hollerith's Tabulating Machine Company called International Business Machines. At the time that the Social Security system was being implemented, IBM was headed by Thomas J. Watson, Sr. Anticipating the problems that would face the Roosevelt administration in setting up Social Security, Watson ordered IBM researchers to begin working on new machines that could handle the enormous workload. By early 1937, researchers had successfully completed work on the IBM 077 Collator, a machine capable of handling and enumerating punch card information at very high speeds. The Collator remained a staple of Social Security operations for the next two decades.

Some anonymous wag once remarked that "to err is human; to really screw things up requires a computer." If that's true, then a lot of the blame can be placed on the Social Security Administration and the Internal Revenue Service for the role that they played in spurring the development of the computer. Those two agencies were not solely responsible, of course: The military was consistently at the forefront of computing innovation, and many large private industries (the telephone company, insurance companies, banks, and so forth) helped underwrite extensive developments in computer technology. Nonetheless, the ongoing, massive data-processing demands of the SSA and the IRS provided computer manufacturers with a steady source of business.

In the decades before the introduction of the commercial mainframe computer, the collection of employee wage information was comparatively benign. The files maintained by the Social Security Administration and the Internal Revenue Service were fairly detailed and extensive, but there was no easy way for the information to be accessed and shared with other government or state agencies. That all changed once government and businesses made the shift from punch cards to electronic bits.

The second major data-processing crunch for the U.S. government occurred just fifteen years after the passage of the Social Security Act, in the early 1950s. By then, the Social Security Administration's punch card files had grown in size to 320 million, outstripping the ability of even IBM's increasingly advanced collating machines to handle the workload.

Fortunately for the Social Security program, the growth in its workload coincided with rapid advances in computer technology, particularly at IBM. In 1952, the company introduced the IBM 701, the first mass-produced computer and the first to use magnetic tape for storing data. Two years later, IBM released the 705 model, which would be the workhorse at the Social Security Administration for years. The Administration took delivery of its first 705 in 1955 (with a whopping 20 Kbytes of memory!), and the advantages of magnetic storage quickly became apparent. Even in those early days of computer development, a single magnetic tape could hold the Social Security records for up to 60,000 people.

[1]The U.S. Employment Service (USES) was created by the passage of the Immigration Act on February 20, 1907. Originally a division of the Bureau of Immigration and Naturalization and later a bureau within the Department of Labor, the USES helped place immigrants in jobs around the country.

[2]USAJOBS website, "Privacy Policy," n.d. Available at www.usajobs.opm.gov/privact.htm.

[3]Garfinkel, Database Nation, pp. 20–21.

[4]United States General Accounting Office, "Identity Theft: Prevalence and Cost Appear to Be Growing," (March 2002). Available on the Web at www.consumer.gov/idtheft/reports/gao-d02363.pdf.

[5]In September 2000, in fact, the House Subcommittee on Government Management, Information, and Technology released its first report card on Computer Security in the Federal Government. In the committee's view, the government's overall grade was a D-, and a number of agencies that routinely gather information about employees—including the Office of Personnel Management, Health & Human Services, Small Business Administration, and Labor—received failing grades.

[6]Ironically, the 1890 census records no longer exist. A portion of the records were damaged by fire and water, and the remainder were destroyed by the Library of Congress in the mid-1930s.

[7]For a fascinating overview of punch card technology and in particular, the history of the phrase "do not fold, spindle, or mutilate," see the article by Steven Lubar, "'Do not fold, spindle, or mutilate': A cultural history of the punch card," Smithsonian Institution (May 1991). Downloaded on June 18, 2002, from ccat.sas.upenn.edu/slubar/fsm.html.

[8]In the early 1960s, the computer industry was jokingly referred to as "Snow White" (IBM) and the "Seven Dwarves" (Burroughs, Control Data, General Electric, Honeywell, NCR, RCA, and Sperry Rand). Today, of course, IBM is better known by the nickname "Big Blue."




The Naked Employee. How Technology Is Compromising Workplace Privacy
Naked Employee, The: How Technology Is Compromising Workplace Privacy
ISBN: 0814471498
EAN: 2147483647
Year: 2003
Pages: 93

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net