Does the Government Manipulate Economic Data?

Americans in this century have been barraged with more economic data than the prior four centuries combined, according to one analysis.

That information overload is fast turning into a form of information pollution. Some economists and experts now say the government consciously and systematically manipulates important economic data.

One of those experts who thinks the US let its economic reality check bounce a long time ago is Kevin Phillips, a political and economic commentator for more than three decades and onetime Nixon strategist.

A generation ago Phillips wrote "The Emerging Republican Majority" which Newsweek magazine described as the "political bible of the Nixon administration," and previously has dissected the fakery in the government deficit numbers-federal, budget, current account and trade deficits-where he says that essentially the country is living on "borrowed prosperity."

Phillips, author of "Bad Money: Reckless Finance, Failed Politics and the Global Crisis of American Capitalism," (Viking, April 2008), says that the government has manipulated economic numbers to create a "make-believe economy" over the last four decades. Phillips says that since the ‘60s, Washington bureaucrats from both sides of the political aisle have pulled the levers behind the scenes to overplay the vitality of the US economy.

Now do I personally expect all data to be perfect? No, it's silly and simplistic to think that way. I also don't think there is any grand conspiracy here. I personally get an allergic skin reaction to the hysterical, hand wringing Emily Litella business journalists whose self-righteous stubbornness and aversion to nuance often drains them of all common sense.

Phillips too is careful to say there is "no vast conspiracy here," just what he calls "Pollyanna creep," a phrase coined by John Williams, a California-based economic analyst and statistician.

"Pollyanna creep" arose from the subtle changes to calculations of economic data that you should be aware of. The methodologies use to arrive at economic data we take for granted as fact are less than exacting. They can produce jarringly wide differences in results-often what Mark Twain once called the difference between a firefly and lightening.

Take inflation. Phillips says the government's manipulation has engendered a fake, low inflation world which fueled low interest rates which unwittingly ignited a debt boom, from colossal government borrowing that's created what I call a Supersize-me government to the junk bond fiasco to the S&L crisis to today's housing and credit bubble. He also says the true economic growth rate, GDP, has been overstated. 

Specifically, Phillips targets three of the most closely watched and what he says are highly manipulated economic measures: The consumer price index, which tracks inflation; the gross domestic product, which tracks the economy's overall growth; and the monthly unemployment figure.

Phillips says that, once you vacuum out the nonsense, inflation is really at 5% (instead of 2%), average annual GDP growth is in the 1% range (instead of the 3% to 4% range), and unemployment is really at 8% (instead of 5%). 

Phillips clearly experiences particular acid reflux for the way the government calculates inflation, notably the fake "core" inflation rate which excludes food and energy costs.

Even the Federal Reserve Bank of Philadelphia and the Wharton School at the University of Pennsylvania, in a research piece within the past two weeks, blasted this fake inflation rate: "We find that food and energy prices are not the most volatile components of inflation and that depending on which inflation measure is used, core inflation is not necessarily the best predictor of total inflation."

And in his surprise statements about the need for a strong dollar last week, Federal Reserve chairman Ben Bernanke gave an unusual argument-he focused on high headline inflation, not core inflation when he noted his concern that rising inflation could directly unsettle inflation expectations, a Fed worry for some time now.

Suspiciously low inflation can dupe investors into buying treasury bonds that in reality have lower real or after-inflation yields. The breakeven inflation rate, or the difference between yields on conventional and inflation-linked bonds, on five-year treasury issues is now 2.4%, the same range for the past four years. But if inflation is really higher, at say 5% annually, then bond investors are getting bamboozled into negative returns.

Phillips raises the specter of this boogeyman. That the federal government is purposely keeping the inflation rate falsely low in order to cut federal payments, from interest on the national debt to cost-of-living outlays for Social Security and disability payments, which are indexed to inflation. If inflation was calculated in the correct manner, Social Security checks would be 70% greater than they currently are, says Williams.

A low inflation rate also makes US GDP look a lot better, as quarterly increases in the dollar value of the country's goods and services get attributed to greater output rather than higher prices.

The government also chucks into the GDP number its best guess of things like how much people can earn from a free checking account, or estimated values of employer-paid health-and life-insurance premiums, Phillips says.

Monkeying around with government data started in the early ‘60s, Phillips says, during the John F. Kennedy administration. It appointed a committee to weigh changes to unemployment data, at a time when unemployment was soaring.

Out-of-work Americans who had quit searching for jobs--even if this was because none could be found--were then labeled "discouraged workers" and excluded from the ranks of the unemployed, though they were previously classified as such, Phillips notes.

In fiscal year 1969, the Johnson administration, with Congress's blessing, orchestrated a "unified budget" that chucked in taxpayers' Social Security funds with the rest of the federal budget, a change that let the government get its mitts on taxpayer Social Security funds for the very first time to use for all sorts of spending programs, including pork barrel projects.

The move, though, masked emerging deficits in Social Security funds, as taxpayer funds that were drawn down were replaced with treasury bonds, essentially more government debt.

Next, President Richard Nixon asked his Federal Reserve chairman Arthur Burns, to concoct a new inflation number that would be split off from traditional headline CPI, dubbed "core" inflation, Phillips says.

This new-fangled "core inflation" would simply knock out, due to nettlesome "volatility," nettlesome food and energy prices. The new number could be shouted from the hilltops and blasted through newspaper headlines whenever the true CPI number was terrifying. It's a number the markets are still too obsessed with today, though some seem to be surfacing out of this delusion.

Let me digress here. Americans don't buy gas and groceries with "core" dollars, as the Wall Street Journal editorial page rightfully notes. And the concoction that is core inflation is most fantabulous when you consider that it includes highly volatile items such as college tuition and health care costs, which are just as rocky, if not more so, than food or energy prices. Why not exclude these items too?

Moreover, the official core inflation stat ascribes nearly the same weight to food and energy as it does to expensive, volatile purchases such as cars and electronics, which consumers buy less frequently. Again, why not exclude all of these items too? Isn't all of this further proof that this is a really fake number?

And doesn't the obsessive focus on core inflation give a dangerously strong bias toward slashing interest rates to record lows, creating more toxic inflation?

When is the Federal Reserve going to be replaced by a computer, as economist Milton Friedman wanted in the '70s?

I do go on. Let me continue with the cooked government data story.

In 1983, Phillips says the Reagan administration monkeyed around even more with inflation data, when the Bureau of Labor Statistics (BLS) decided that housing, too, was overstating CPI.

So, the BLS swapped in what it calls an "owner equivalent rent" measurement, what homeowners would pay to live in their homes if they were renters. But that number likely understated housing costs as it is based on overall rent, which stayed flat in most of the country during the housing bubble.

So, the government has cooked up its own housing inflation number that likely understates home prices, Phillips argues, and in turn has understated housing inflation during the recent housing boom by three to four percentage points.

Moreover, Phillips says in the 1990s, the CPI has been subjected to three other adjustments, all delivering a downward bias and all dubious:

*Product substitution: If flank steak gets too expensive, people are assumed to shift to hamburger, but nobody is assumed to move up to filet mignon, he says;

*Geometric weighting: Goods and services in which costs are rising most rapidly get a lower weighting for a presumed reduction in consumption

*And, most strangely, hedonic adjustment: An unusual bit of monkeyshines by which the government says that product improvements in things like computers, cell phones or television actually amount to a reduction in price, so a $2000 laptop with a built in camera is less expensive than a $1500 laptop without one. 

Pollyanna creep in the inflation data continued under the Bush administration. In 2006 it stopped publishing the M-3 money supply numbers, which captured rising inflationary impetus from bank credit activity, Phillips says.  

Under the Clintons, Phillips says, the nation's employment figures were massaged and kneaded too.

In 1994, the Bureau of Labor Statistics redefined the work force to include only that small percentage of what it called "discouraged workers" who had been seeking work for less than a year, Phillips says. The longer-term "discouraged"-some 4m U.S. adults who simply are not working-fell out of the main monthly tally. Some now call them the "hidden unemployed."

The Clinton administration also dropped the number of households sampled for the data, from 60,000 to 50,000, making the number more rickety.

But a disproportionate number of the dropped households were in the inner cities. So, along with a new adjustment formula that is believed to also have cut black unemployment estimates, poverty figures get to look a lot less worse, Phillips says.

As for the way gross domestic product is manipulated, just check out one method the government uses to arrive at this number, the adjustments made to calculate the launches of new businesses and the ending of old ones, Phillips notes.

Since the government can't pay for bureaucrats to roam every market in the land, a calculation does it for them. Watch how this plays out in the data the government reported that said payrolls shrank by just 20,000 jobs in April, and that the unemployment rate dropped to 5% from 5.1%.

Check this out. Recent job losses would have approached 287,000 if not for the government's use of the so-called ‘birth-death' model, which adjusts jobs added or lost by estimating the number of businesses that either launched or were shuttered each month.

By this measure, the financial services sector added 8,000 jobs in April at a time when its estimated 36,000 have already been laid off, with an estimated 24,000 more layoffs coming. The government then suddenly said the economy lost 3,700 financial services jobs. It said too that in April the construction sector added, get this, 45,000 jobs. Then suddenly in May it said construction shed 34,000 jobs.

Take a dramamine.