|
The data for most series in Historical Statistics of the United States are expressed in American units. This system of weights and measures was adapted and modified over time from the British Imperial System and is now known as the U.S. Customary System. Table Ap-A presents the official relationships within the U.S. Customary System. It should be noted that the British Imperial System and the U.S. system share many common terms (for example, "gallon” and "bushel”), but the measures are not always equivalent. At this time, only three countries – Burma (Myanmar), Liberia, and the United States – have not adopted the International System of Units (SI, or metric system) as their official system of weights and measures. The SI (for Système International, the French name) was adopted in 1960 by the Eleventh General Conference on Weights and Measures. The key features of the International System are decimalization, a system of prefixes, and a standard defined in terms of invariable physical measures. Table Ap-B presents the standard system of prefixes used in the metric system. The first U.S. Secretary of State, Thomas Jefferson, influenced by a proposal circulating in France, proposed a decimal system for weights, measures, and coinage of the United States in 1790. However, only his proposal for the coinage was adopted when two years later the U.S. mint introduced the world's first decimal coinage (a dollar of 100 cents). France adopted the decimal metric system in 1795. An international treaty signed by the United States in 1875 established the metric system as an international standard.
The U.S. Customary System has many drawbacks: its nondecimal nature makes it more complex to convert from one unit to another; it uses the same name for different units (e.g., ounce for both weight and liquid capacity, quart and pint for both liquid and dry capacity); and it has three different systems of weights (avoirdupois, troy, and apothecary). Although U.S. law has sanctioned the use of the metric system since 1866, SI has yet to displace the customary system in everyday use. While the United States does not use the metric system in its commercial activities, there is increasing acceptance of the metric system in science, medicine, government, and many sectors of industry. Important equivalencies are provided in Table Ap-C as an aid to users who might wish to convert data in this volume into the International System of weights and measures.
Table Ap-D provides more extensive information on units of measure for agricultural products, which have been sold and transported in a bewildering variety of different shipping containers. Before the middle years of the nineteenth century, most products were sold by volume, not weight. Article I, Section 8 of the U.S. Constitution gave Congress the power to establish legal weights and measures, but in practice the legal definition of these measures and the size of containers were left to the states until the late nineteenth century. It was not uncommon for different states to define a legal measure differently and to define a measure differently when used for different products. As farmers began to have access to weight scales and as products were sold more regularly over long distances, sales increasingly were based on weight. Tradition, however, dictated that quantities and prices continued to be denominated in volume measures (for example, bushels of corn). Conversion factors were used to equate volume measures and weight. At first these conversion factors were informal rules of thumb, but states began to legislate the legal weight of various volumes, crop by crop. Gradually over the nineteenth century, national standards began to emerge for most crops. Generally speaking, dry products are measured by bushels and liquid products are measured by gallons. The standard bushel that evolved in the United States is formally known as the "Winchester bushel” (see Table Ap-A).
Note that, in some cases, units of measure for particular data series may differ from the standard values presented in Table Ap-A and Table Ap-D. As always, users should consult the table documentation.
Many of the series in Historical Statistics of the United States give the (average) price of an item or the value of an aggregate such as total expenditure or total value of production. Generally speaking, these values are measured either in market prices of the year to which the data refer or in "real dollars,” which are sometimes called "constant dollars.” In the former case, the interpretation of the data is usually straightforward. The actual price or value of the item in the year it was sold or produced is presented in U.S. dollars. These are called current prices, market prices, or values in current dollars or nominal dollars. These terms are interchangeable. However, users should take notice of the specifics provided in the table documentation for the series in which they are interested.
From year to year, the value of the U.S. dollar has fluctuated, and over time there has been a general tendency for the value of the dollar to fall. It is often desired to make comparisons of the value of some item over time. Consider, for example, the Ford Motor Company, which was incorporated in 1903 by Henry Ford. In October 1908, Ford offered his Model-T for $850. What would that be equivalent to today? In 1914, Ford began paying his employees five dollars a day, nearly doubling the wages offered by other manufacturers. How much is that in today's terms?
There is no universally accepted answer to these questions. 1 Determining the relative value of an amount of money in one year compared with another is a matter of often-subtle historical judgment, and the appropriate method will depend upon the context of the question. The value of the dollar fluctuates because prices fluctuate. If, on balance, prices rise, the value of the dollar falls. Of course, all prices do not rise or fall in lock-step with one another. The most common method of calculating an equivalent price is to use a price index that represents an average of the prices comprising a bundle of goods and services (see Chapter Cc). A common choice is the Consumer Price Index (CPI), which averages prices paid by urban consumers for a representative basket of goods and services ( series Cc1). Price indexes such as the CPI are set equal to 100 in a base reference period. In this case, the base years when the index averages 100 are 1982–1984. In 1908, the year of the first Model-T Ford, the CPI was 9.235. In 2001, the index number was 177.100 ( series Cc1). Thus, average prices as measured by this index rose more than 19-fold (177.100/9.235 = 19.18). The $850 asking price is equivalent to $16,300 in 2001 (19.18 × 850 = 16,303). Before accepting this result, there are several points to consider. First, the CPI is only one of many price indexes we might have chosen. Because any price index is only an average of prices for a finite list of goods, indexes will differ from each other, often widely. The CPI uses a list of goods and services purchased in 1993–1995 by urban consumers. Thus, the comparison just calculated represents the value of the Model-T in terms of the consumption goods included in the index. Another alternative would be to use a producer price index that represents the price of inputs that an automobile manufacturer might purchase. A third alternative would be to use a broad index such as the gross domestic product (GDP) deflator (see Table Ca149–158). Second, when making such calculations over a relatively long time span, the representative bundle of goods will change, both in its make-up and in the inherent quality of the goods and services that are included. In practice, the goods included on the list are changed periodically, and the resulting averages are "spliced” together to form a long continuous series. The automobile purchaser of 1908 could not have purchased many of the goods and services on the 1993–1995 CPI list (frozen orange juice, television sets, modern medicines). You might ask, would you rather have $16,300 that you could spend on items obtainable in 2001 or $850 to spend on items obtainable in 1908? Which is a "better” car, the 1908 Model-T or a $16,300 Ford manufactured in 2001? Interpreting these converted prices is tricky business, and the further one goes into the past, the trickier the task becomes.
An alternative to using a price index to compare values over time is to compare the historical price with the typical wage of a worker at the same time or with per capita income. At the $5 for an 8-hour day that Ford offered workers in 1914, they would have had to work for 170 days before earning $850. Before this raise, the majority of workers had been earning $2.34 per 9-hour day (Raff 1988). At that rate, over 360 days of work would be required to equal the Model-T. In 2001, a worker earning the federal minimum wage ($5.15 per hour) would make only $7,000 (before tax) in 170 days and $15,000 in 360 days, still not enough to cover the price of a $16,300 automobile.
Begun in 1817, the Erie Canal was completed in 1825 at an approximate cost of $7 million. This waterway is regarded as one of the most important transportation investments of the nineteenth century (see Chapter Df). How does its cost compare with transportation projects of recent times? Using the CPI, it would be about $100 million, sufficient for only a few miles of interstate highway.
This calculation, however, trivializes the vast size of the undertaking represented by the Erie Canal project. An alternative to using a price index is to estimate the fraction of GDP accounted for by the project. Our estimate of GDP for 1825 is about $1 billion ( series Ca10); thus, the canal cost 0.7 percent of an entire year's output. In 2001, GDP was more than $10 trillion, and 0.7 percent of that amount is $70 billion. As a comparison, the fiscal year 2001 budget of the U.S. Department of Transportation was approximately $60 billion.
Raff, Daniel M. G. 1988. "Wage Determination Theory and the Five-Dollar Day at Ford.” Journal of Economic History 48 (June): 387–99.
Williamson, Samuel H. 2002. "What Is the Relative Value?” Economic History Services, at the EH.net Internet site.
......................................
| 1. |
A useful resource is Williamson (2002). Several of the examples given later in this appendix are from this source.
|
|
Top 
