Comments Off

Michael Boskin is a professor of economics at Stanford University and a senior fellow at the Hoover Institution. Today, in the Wall Street Journal, he proposes a temporary payroll tax reduction as a second try at a stimulus. He starts his argument with consideration of what would have happened with a temporary payroll tax reduction in February of this year:

My Stanford colleague Pete Klenow and Rochester economist Mark Bils estimated that cutting the payroll tax by six percentage points (of the 12.4% Social Security component) would, under standard assumptions, increase employment by three million to four million workers—an amount equal to all the job losses since the stimulus was passed.

The payroll tax cut would have reduced firms’ costs by roughly the same amount as from the entire decline in employment. It would have cost less than half as much as the stimulus bill, gotten far more income into paychecks quickly and, most importantly, greatly reduced incentives for firms to lay off workers. In fact, it would have created incentives to hire.

Instead, we choose a stimulus that critics called “porkapalooza”, and have seen unemployment exceed targets by about 25% (to over 10% instead of the promised maximum of 8.6%).

Boskin’s recommendation is a “partial payroll tax reduction”, cutting the tax in half for a few months. That means the average wage earner would see a 3% raise for those months (and the employer would also see a 3% drop in payroll taxes). On the employer side, that might spur hiring or help employers retain employees slated for layoffs, but I don’t think its enough of an effect for the wage earner.

Boskin also recommends accelerating some spending “that has to be done anyway”, such as replenishment of military stores. I have my own reservations about opening the door to any more government spending. The level of military spending is certainly something open to debate, anyway, and the current administration is more likely to try to cut military spending after huge increases for the war in Iraq and Afghanistan.

In my view, a better idea would have been a three-month payroll tax “holiday”, with workers enjoying an increase in wages equal to 6%, and employers enjoying an immediate 25% reduction in taxes for the first quarter of next year.

Unlike income taxes, payroll taxes are levied disproportionally on the lower and middle income classes, with every dollar of their income taxed. The Earned Income Tax Credit is designed to offset some of that disproportionate burden, but it is reduced and eventually phased out for those earning over $43,279. People in the third and fourth quintile of income pay more of this tax as a percentage of total income than any other class. Because the payroll tax is collected on every dollar up to $106,800 in income, the richest families see payroll tax free income above that level. (All rates are 2009 figures.)

The middle class is the “engine” of consumer spending, and a reliable, defined 6% increase in wages for three months would do much to spur delayed spending on consumer goods. And it would be “kinder” to our budget deficit than “stimulative” spending.

While taxes on the “rich” are often decried by my fellow conservatives, when all taxes and tax credits are added together our system looks moderately progressive and not “confiscatory”. Economists separate household income levels into 5 equally sized groups, or “quintiles”, with income and effective tax rates per quintile, as shown:

Quintile Income Bracket Source: http://www.bls.gov/cex/2006/Standard/quintile.pdf

These are the estimated percentages actually paid in 2006, not what the tax tables state. Note that “the rich” don’t pay 39% on every dollar of income; with lower taxes on the income up to their top level, tax deductions and other credits, they end up paying 25.8% of their income. It would be simpler to incorporate these tax rates into a tax code of a single page instead of the 70,000+ page tax code we now have, but that’s another issue entirely.

The tax percentages above include payroll, income, excise and other federal taxes, minus deductions and credits, as estimated by the Cato Institute. The income levels for the quintiles, from the BLS, are “household” income (two wage earners if filing jointly). Many are shocked to find their family is in the 4th or 5th quintile, but you are in the top 40% of wage earners if your combined income is over $56,222. If you and your spouse earn $89,000 combined, you are “rich”, in the top 5th of all income earners.

Unlike the Making Work Pay tax credit, passed with the 2009 stimulus bill, a payroll tax holiday would not be accompanied by a temporary adjustment in the withholding tables that might “bite” some taxpayers. The $400 per person “Making Work Pay” tax credit was offset by automatic deductions in the withholding tables, providing the average wage earner with a few extra dollars each paycheck. But that affects the total amount withheld, and some tax payers may see a tax bill due instead of a refund at tax time in the spring. That tax surprise will not be pleasant for families already pressed by salary reductions, rising credit card interest and other effects of the recession.

Rep. Louie Gohmert had proposed a payroll tax holiday as an alternative to multi-national corporate bailouts in late 2008, but the proposal went nowhere. It is time to reexamine this proposal, perhaps even on a bi-partisan basis – one can hope – for the first quarter of 2010.

Cross-posted to Donklephant.com

Comments Off

The administration’s claim of nearly 640,000 jobs created or “saved” by the $787 billion stimulus bill continues to be savaged by traditional media outlets for inaccuracies.

From the Washington Examiner:

More than ten percent of the jobs the Obama administration has claimed were “created or saved” by the $787 billion stimulus package are doubtful or imaginary, according to reports compiled from eleven major newspapers and the Associated Press.

Based only on our analysis of stimulus media coverage in the last two weeks, The Examiner has created this interactive map to document exaggerated stimulus claims. The map, which will be updated as new revelations appear, currently reflects an exaggeration by the Obama administration of about 75,000 jobs, out of the 640,000 jobs supposedly “created or saved.”

From ABC News, a caution about some of the details:

Here’s a stimulus success story: In Arizona’s 15th congressional district, 30 jobs have been saved or created with just $761,420 in federal stimulus spending. At least that’s what the Web site set up by the Obama administration to track the $787 billion stimulus says.

There’s one problem, though: There is no 15th congressional district in Arizona; the state has only eight districts.

And ABC News has found many more entries for projects like this in places that are incorrectly identified.

And the Sacramento Bee notes some problems with the statistics from the Golden State:

Up to one-fourth of the 110,000 jobs reported as saved by federal stimulus money in California probably never were in danger, a Bee review has found.

It would be easy to criticize the administration for lacking the critical oversight necessary to get the statistics right. After all, none other than VP “Nobody Messes with Joe” Biden was assigned to be the official watchdog.

But piddling little details like actual real numbers aside, economics is the dismal science. And its very hard to determine exactly what effect a spending program has on the job market.

The Wall Street Journal, usually not considered a bastion of liberal economic thought, observes that:

It’s easy to ridicule the White House’s estimate that 650,000 jobs have so far been “saved or created” by the $150 billion spent so far from their $787 billion stimulus bill. That’s because while it’s possible to measure jobs created by the stimulus (for example, counting the number of construction workers on an infrastructure project funded by the bill), it’s a lot less clear how you measure jobs saved by it.

“One can search economic textbooks forever without finding a concept called ‘jobs saved’,” said Allan Meltzer, a professor at Carnegie Mellon University, in a memo to House of Representatives Republican leader John Boehner today. “How can anyone know that his or her job has been saved?”

The disparate estimates come down to the multiplier effect. There are jobs directly created, and then there are jobs that are created or saved because the direct beneficiaries are still employed. For example, the stimulus money may have financed a construction workers’ job, but it also may have saved the job of someone at the grocery store where he shops. There’s also the issue of government jobs saved. Many states were planning layoffs that were either canceled or postponed because of stimulus money. But again, those numbers are difficult to quantify.

The Journal’s panel of 46 economists had predicted a loss of 271,000 jobs per month without the stimulus bill, and a less-catastrophic loss of 183,000 jobs if the bill passed. Reality, as if to prove the “dismal science” label correct, chimed in with a job loss of about 400,000 per month with the stimulus.

That panel’s prediction was the basis for VP Biden’s claim that the stimulus will create or “save” a million jobs, according to the Journal.

There’s an old joke that goes like this: If you laid all the economists in the world end to end, they still wouldn’t come to the right conclusion.

And that’s the danger in relying on government to solve economic problems. You hope the predictions are right before the biggest entity in the known universe sucks all the oxygen out of the room.

Cross posted to Donklephant.com