Robert Samuelson April 16, 2014

April 16, 2014
By Robert Samuelson

April 16, 2014
 


 In this age of Big Data, the Obama administration seems determined to make it hard for average Americans to get little data. Recall that in2011, the Census Bureau decided to eliminate “The Statistical Abstract of the United States,” first published in 1878. This was the nation’s best compilation offigures on hundreds of topics, ranging from birth rates to forest firesto wage rates to voting patterns. The Census Bureau said that itcouldn’t afford the Stat Abstract (the savings were trivial, about $3million) and that most figures were online.

True. But this doesn’t mean that people can get them easily.You have to know where online to go. (The sources were dozens ofgovernment agencies, many private industry groups and surveyorganizations.) You also have to know how to manipulate the databases to retrieve the desired statistics. Overcoming these problems oftenrequires a large investment of time and much frustration.

The lesson wasn’t learned.

The president’s Council of EconomicAdvisers (CEA) has now committed the same blunder. For decades, theannual report of the CEA has contained an enormously valuablestatistical annex of about 100 tables; in the 2013 report, there were 112. In 45 years of reporting, I have relied on the CEA’sstatistical annex as a basic source of economic information. Nothingelse approached its ease of use, clarity and length of the statisticalseries, often covering four or five decades. The latest CEA report eliminates three-quarters of the tables. It calls this purge a “streamlining” andoffers the same explanation: The data are online.

Here are sometables that got the ax: historical stock market prices; farm output andproductivity; average maturity of Treasury debt securities; labor forceparticipation rates by demographic group; national income by type ofincome (wages, salaries, profits, rents, etc.); U.S. trade by geographic markets. The list goes on.

Again, Web sites are not goodsubstitutes for the old-fashioned tables. True, the CEA usefully printssources for all the eliminated tables. The implication is that it’s asnap to duplicate them from the Internet. Not so.

Take the firsttable on my list above: stock prices. For starters, there are foursources for the table: the New York Stock Exchange; Dow Jones & Co.; Standard & Poor’s Financial Services; and the Nasdaq Stock Market.Beginning with the New York Stock Exchange, I type https://nyse.nyx.com and get its home page. There’s no obvious link to data back to the1960s and earlier. I click on “NYSE composite,” the desired price index. This provides today’s index move. Again, no obvious link tohistoric data. It’s not clear how long it would take to reconstruct atable that the CEA had routinely produced.

To be fair, the CEAfaces budget pressures (it says its budget is down about 20 percent from the peak) and some historical statistical series are revised afterbeing printed in the annual report. Also, the CEA has a monthlypublication, “Economic Indicators,” that provides data going back a decade for some of the statistics. Butthese are footnotes that do not justify the CEA’s wholesale statisticalslaughter.

The remedy here is obvious: Undo the deed. Given human nature — which loathes admitting error — this seems unlikely.