It is time to reveal a dirty little insider's secret. It isn't hard to make time-series forecasting models produce wildly different results.
Estimates have a tendency to whizz around all over the place in the face of small changes in assumptions, data used and time period covered. And errors can be huge, especially at turning points, so carefully considered judgment matters. Economic forecasts can be massaged, so independence is vital.
During my time on the Bank of England's Monetary Policy Committee, which makes quarterly economic prognoses, Governor Mervyn King controlled the hiring and firing of the forecast team, who did his bidding. They had to produce a result that was consistent with King's views, or else they would be history. A patchwork of arbitrary fixes and prejudices frequently drive forecasts, which for the uninitiated are hard to see.
King always emphasized the importance of top-down judgments, which means you can just make stuff up as you go along. Worryingly, this was often only loosely based on the workings of the real world. Such glorified guesswork operated reasonably well during the boom years, but failed miserably when the recession hit. To put it bluntly, it isn't that hard to manipulate a forecast. I have seen it done.
On May 17, 2010, the UK's new Chancellor of the Exchequer, George Osborne, announced that the Treasury would stop producing its own economic forecasts.
"I am the first chancellor to remove the temptation to fiddle the figures by giving up control over the economic and fiscal forecast," he said. "I recognize that this will create a rod for my back down the line, and for the backs of future chancellors. That is the whole point. We need to fix the budget to fit the figures, not fix the figures to fit the budget. To do this, I am today establishing a new independent Office for Budget Responsibility (OBR). For the first time, we will have a truly independent assessment of the state of the nation's finances."
Bold stuff. Unfortunately, after only a couple of months the OBR is proving to be neither responsible nor independent. Instead, there is growing evidence that this new organization had actually fiddled its own figures to get the new coalition government out of a jam. The OBR, as currently constituted, has failed almost as soon as it began.
An office with "responsibility" in its title and "independent" in its Web address was always going to have a lot to live up to. The OBR has manifestly shown that it isn't taking care of its duties and doesn't appear to be autonomous. Appearance is everything.
The government has claimed the OBR is independent even though it is physically located in the Treasury; it is staffed by a handful of seconded Treasury officials; all queries are handled by the Treasury press office; and if you call them, the Treasury switchboard answers.
On June 22, Osborne announced a program of measures to cut public spending and raise taxes, including an increase in the value-added tax rate to 20 percent from 17.5 percent. The OBR produced a forecast that suggested these measures would have a minimal effect on employment and unemployment.
Subsequently, leaked documents showed they would actually result in the loss of at least 500,000 public-sector jobs and 600,000 to 700,000 jobs in the private sector.
By noon the next day, the OBR had produced a new forecast saying that, even with these job cuts, employment would rise and unemployment would fall every year throughout the forecast period. The idea that the private sector would fill this hole by creating as many as 2.5 million jobs was greeted with howls of incredulity. And with a new forecast produced so quickly, it smacked strongly of political interference.
Adding to the firestorm, Alan Budd, the OBR's boss, almost immediately announced he would be leaving. The government's claims that Budd's appointment was always intended to be temporary don't appear credible.
A few days later, the plot thickened when it was reported that the OBR had also put a positive gloss on the employment numbers by trimming its forecasts for public-sector job losses by about 175,000.
The OBR pre-empted the results of the Pensions Commission by assuming lower pension contributions and reduced promotions for public servants, even though the government hasn't announced such a plan. Both assumptions cut the job-loss figure. Meanwhile, policy initiatives that would lower long-term growth and increase unemployment were excluded.
As a young economist, I gave a presentation on youth unemployment to an academic panel at the Congressional Budget Office, which produces a forecast twice a year for the US Congress. The panel's purpose is to "provide advice to further the reliability, professional quality and transparency of the CBO's work."
Before the talk, I had no idea who was on the panel. Sitting in front of me, among others, were Nobel laureates Lawrence Klein, Paul Samuelson, James Tobin and Robert Solow, who had to sign off on the forecast before it was made public. That is independence.
David Blanchflower, a former member of the Bank of England's Monetary Policy Committee, is professor of economics at Dartmouth College and the University of Stirling. The opinions expressed are his own.
Source: China Daily
Our IP Address: