I joined a large number of economists signing an open letter supporting funding for economic data. The letter is here, twitter #SaveTheData, Financial Times story here, press release here.
Few public goods are as cheap or important as good economic data. Much of our national policy discussion is based on government-collected data. Changes in inequality, wage growth or stagnation, employment and unemployment, growth, inflation... none of these are readily visible walking down the street.
Free, openly accessible, well-documented data, allowing comparisons over long periods of time, such as provided by the Bureau of Labor Statistics, is especially valuable.
Already, much of the data we get is based on decades-old measurement concepts. Perhaps someday internet big data will bring us alternatives. But that day is a long way away. Let's not fly blind in the meantime.
Imagine the price of a space telescope, a particle-accelerator or on a lower scale an MRI machine. All with the purpose of making better or new observations. Data collection is the sensor of social science, probably (my guess) a lot cheaper than most natural science measurement.
ReplyDeleteWhat's the evidence that the social marginal benefit of more data is >0? It's a serious question. For my research, I want more data. But is there evidence it improves national policy?
ReplyDeleteIn the social sciences there is something called the Hawthorne effect:
Deletehttps://en.wikipedia.org/wiki/Hawthorne_effect
I bombard you with questions regarding your opinion of a certain good (for instance a Ford truck) and you gradually take interest in the good I am asking you about. And the data collection process itself can be turned into a form of economic policy. Advertisements disguised as questionnaires is the oldest trick in the advertising book (see Coke versus Pepsi challenge).
It is very difficult in the social sciences to conduct a double blind experiment.
Finally, this statement by John is a real hoot:
"Free, openly accessible, well-documented data, allowing comparisons over long periods of time, such as provided by the Bureau of Labor Statistics, is especially valuable."
People that work at the BLS don't do it for free. Ultimately the cost of data collection is born by the taxpayer.
The British governor of Hong Kong, who was an advocate of free markets, would not authorize the collection of economic data, because if the bureaucrats have data about something, they will want to control it. I think that we should reduce the amount of economic and social data that the US government collects.
ReplyDeleteJohn, your support for this is very disappointing. Just another special interest group fighting at the teat of government. Heaven forbid academics find creative voluntary ways of "measuring" economic factors.
ReplyDeleteI have exactly the opposite reaction. I'm extremely happy to see this posting by Cochrane, for two reasons: (1) supports the importance of empirical evidence in macroeconomic argument, and (2) demonstrates flexibility and open mindedness to a degree I've not often seen here. Bravo.
DeleteAs I would have expected, several of the early comments to John's post were negative. I'm not sure whether the negative commenters wanted simply to keep the statistical budgets declining in real terms or they wanted to eliminate the statistical programs altogether. Either way, I think the commenters are on to something. Perhaps cutting statistical programs could be expanded to the military, by reducing or eliminating the budget lines for intelligence. If there are objections to such a proposal by people who think that the military needs to know what is going on, perhaps we should privatize military intelligence collection. Turning the job over to Halliburton sounds like a real winner.
DeleteGee, what a surprise... a group of economists signs a petition asking for yet more taxpayer support of their "profession"!
ReplyDeleteAt least I'm glad that a few other regular readers (especially Paul and, I believe, Frank and Fat Man) are also disappointed in Cochran.
As for those (such as "Anonymous") who, in their typical elitist/snobbish way (I'm so impressed and intimidated by someone who doesn't even have the courage to give his name!), insist that more taxpayer subsidy is needed, let me assure you that most working class people do not need economists to tell us what the politicians want us to believe about how great the economy is doing, when we know otherwise simply by talking with our neighbors, friends, and relatives.
What we need, rather, is for the government to stop being the central bank (as it has been since about 1913) and otherwise meddling in the monetary system. A position that Cochran has (I believe) supported in the past (I know certainly that Ron Paul and many others support Ending the Fed), making his signing this proclamation rather puzzling...
could it be that it has never occurred to him that if it weren't for all this government control over the banking/monetary system, that 90% of economists would be out of a job (since 90% of them are employed by the government to justify its manipulations, or by those trying to predict the consequences of those government manipulations)?!
I used to believe in what you say, until I researched the evidence: money is neutral (Google this). A starting point is Wikipedia. A paper of interest is Ben S. Bernanke's 2003 FAVAR paper (Google it). The "Fed" has about 3.2% to 13.2% influence on a range of variables, out of 100%, that is, nearly nothing. Bernanke is not kidding you when he testified before Congress that the Fed is not that powerful. He was not being modest, he was probably thinking of his earlier econometric paper.
DeleteNeal,
Delete"What we need, rather, is for the government to stop being the central bank (as it has been since about 1913) and otherwise meddling in the monetary system."
Prior to the establishment of the Federal Reserve (1913) the U. S. government has provided a central bank in the form of the First Bank of the United States (1791-1811) and Second Bank of the United States (1816-1836).
As for government "meddling": Article I of the Constitution
https://en.wikipedia.org/wiki/Article_One_of_the_United_States_Constitution#Enumerated_powers
The Congress shall have power:
To borrow Money on the credit of the United States
To coin Money, regulate the Value thereof, and of foreign Coin, and fix the Standard of Weights and Measures
The Constitution grants government fairly wide latitude in monetary "meddling".
The U. S. Federal Reserve (established in 1913) was created as a compromise to provide the United States a single unified nationwide currency (not a bad thing to have), to establish a system of payments between banks (again not a bad thing), and to provide a lender of last resort (some may argue against such authority).
Open market operations by the U. S. central bank were not permitted until the creation of the FOMC 20 years later with the Banking Act of 1933 (aka Glass-Steagall).
The funny thing is that Alan Greenspan was a leading proponent for eliminating several provisions of that act (including the separation of commercial and investment banking), but the FOMC was kept intact.
"...could it be that it has never occurred to him that if it weren't for all this government control over the banking/monetary system, that 90% of economists would be out of a job..."
Ask Mr. Greenspan - was retaining the FOMC provision of the Glass Steagall Act a means to "save jobs" for economists in government employ.
Prof. Cochrane is saying the statistics are worth the cost in taxes. He brings some expertise to the table. In an indirect way, he's also saying some government functions are useful. It isn't as if he suddenly turned into Bernie Sanders or PK. The negative comments are extreme.
ReplyDeleteYour writing is very interesting and educational, but I tried to go to "next blog" and my computer security stopped me saying it's not a safe webpage. Just thought I'd let you know so if there's something you can do about it .....
ReplyDeleteGene Epstein, the economist, wrote about this decades ago and I think 'owns' this issue. Scott Sumner has blogged on the need for a NGDP futures market. iPredict in New Zealand has tried to set up various prediction market events but has failed due to lack of interest. Signing an open letter to some government bureaucrat will send the bureaucrats at some level scrambling for a response, but is otherwise ineffective. Unless and until the free market is interested in this data, perhaps for 'investment / gambling' purposes, it's a dead letter.
ReplyDelete"Already, much of the data we get is based on decades-old measurement concepts. Perhaps someday internet big data will bring us alternatives. But that day is a long way away. Let's not fly blind in the meantime."
ReplyDeleteThere is an inherent inconsistency here. Professor Cochrane is absolutely correct that most government-driven economic policy depends on accurate and more-or-less real-time data. While it is not a given that improved data would improve policy (but perhaps only ill-advised policy implementation---much of this is politics), it is an appropriate, albeit much too rare, admission that the data collection apparatus is not currently sufficient for the purposes it is now being used. Until it is, economists should humbly back away from their predictions and get away from the economic cockpit. We are currently flying blind. There is no such thing as "we should not fly blind in the meantime".