The Sharp Pencil Test
Time for a real growth agenda
Jun 13, 2011, Vol. 16, No. 37 • By LAWRENCE B. LINDSEY
It’s easy to get caught up in the details of the political battle of the day over the nation’s economic and fiscal health—after all, that’s what we do in Washington. Unfortunately, many of our decision-makers and opinion leaders possess the skills required for political infighting in greater abundance than the capacity for thoughtful analysis. Their focus on the fight and on winning the 24-hour news cycle has produced a country that designs policy for tactical political advantage and not for the efficient use of scarce resources. As a consequence we have dug a deep hole that will be difficult to escape from. As bad as things may seem, our country’s actual long-term economic and fiscal situation is even worse than Washington’s political class recognizes.
Start with economic growth. The president’s budget assumed 3.1 percent growth this year and 4 percent next year. It now looks like the first half of this year will produce a number just over 2 percent, and there is no evidence that growth is going to accelerate. Since the recession formally ended in the second quarter of 2009, growth has averaged just 2.8 percent, a little over half the average of 4.6 percent in the two years following the typical recession since World War II. The reasons are partly structural, but many government policies aggravate the economy’s structural weaknesses. The effect on the budget can be profound. Each percentage point of slower real growth increases the cumulative budget deficit by $3.2 trillion over ten years. Even if growth meets current projections, federal debt per capita will rise from $19,000 in 2008 to $68,000 in 2020.
Then there is the problem of interest rates. Right now, thanks in large part to Federal Reserve policy, Uncle Sam can borrow at an average cost of just 2.5 percent. The average borrowing cost over the last three decades was 5.7 percent. Our debt is now $14 trillion and scheduled to grow to $25 trillion by the end of the decade. If interest rates normalize over that period, the added interest costs in 2021 alone will be $800 billion—more than 20 times the mere $37 billion in budget cuts that tore up Congress in March. It would take virtually all of the cuts in the Ryan budget just to cover that added interest, much less to start bringing down the national debt. Unfortunately, the Fed is now in a fiscal box. A normalization of interest rates would break the Treasury. Hence, a normalization of rates really can’t happen—we are stuck in a world in which the Fed must keep rates artificially low in order to prevent a budget disaster.
Government policies to “stimulate” growth have not done so. Everyone except flacks for the White House knows that the 2009 stimulus package failed miserably to produce the promised results. But even if you buy the White House’s argument that the $800 billion package created 3 million jobs, that works out to $266,000 per job. Taxing or borrowing $266,000 from the private sector to create a single job is simply not a cost effective way of putting America back to work. The long-term debt burden of that $266,000 swamps any benefit that the single job created might provide.
This is an example of a program failing the Sharp Pencil Test. If you sit down and do a back of the envelope calculation of the program’s costs and benefits, there is no way to conjure up numbers that allow it to make sense. But the stimulus bill is hardly alone. What we need to do is take a sharp pencil to all of our programs—and to our tax code—and redesign them in a way that brings maximum benefit at minimum cost.
For a case study in the failure of political program design, consider Obamacare. It is widely, if privately, acknowledged across the political spectrum that Obamacare will have to be reformed before it takes effect in 2014. Recall how it became law. On the day before Congress’s Christmas break in 2009, Senate majority leader Harry Reid called a bill to the floor that had been cobbled together behind closed doors with last-minute agreements penciled in the margins. Several Democratic senators stood and said they thought the bill wouldn’t work, but they were going to vote for it in order to allow it to move to a conference committee with the House, where the inconsistencies of complex bills are typically ironed out. But before the conference committee could do its work, Scott Brown won the Massachusetts Senate election, and there were no longer the 60 votes the Senate would need to pass the conference bill. So the House had to accept the Senate bill virtually unchanged. To get over thorny issues like abortion funding, Obama had to sign an executive order since the bill couldn’t be reopened.
Ideally Congress would have started over and designed a workable program, but it did not. Secretary of Health and Human Services Kathleen Sebelius has been trying to stitch the unworkable result together by granting waivers—1,400 of them already—to such Mom and Pop businesses as McDonald’s to stop them from dropping their health benefits. A joint study by the departments of Labor, Treasury, and Health and Human Services found that a majority of private sector health plans would not comply with the rules HHS was drafting. When firms face a choice between complying with these regulations and putting their workers on the public plan at a fraction of the cost, the whole budgetary house of cards will collapse.
But there is no room for partisan smugness. A number of Bush administration ideas also fail the Sharp Pencil Test. Ethanol subsidies cost far more than any rational computation would justify. There is no evidence that No Child Left Behind has significantly improved test scores, though it has prompted the dumbing-down of tests. And the bipartisan establishment’s immigration reform developed under Bush—making 11 million people citizens—is so obviously beyond the capacity of the already clogged Immigration and Naturalization Service as to fail the laugh test. (See my “Can Immigration Reform Work?” in the May 22, 2006, issue.)
But probably the hardest thing for Republicans to admit is that the nation-building part of the liberation of Iraq and Afghanistan has not been cost effective. One can easily make a case that deposing Saddam and the Taliban was a cost effective use of the nation’s resources. One cannot make the case that being there eight years later is. While nation-building is a nice idea, the trillion-dollar price tag for Iraq and Afghanistan is too high. President Obama’s decision to do regime change on the cheap in Libya is unlikely to prove any more successful than Secretary Rumsfeld’s effort to topple Saddam on a budget of $54 billion in 2003.
Each of the major components of federal spending requires a structural repair, and so does our tax policy. Currently the focus is on discretionary spending—which is scheduled to total about $16 trillion over the decade, about evenly divided between defense and non-defense. Obviously government wastes a lot of money in this category. We need to be candid about the fact that there are some things we simply cannot afford to do. These uneconomic programs range from dozens of low cost initiatives to some fairly grand ones. No, we do not need to subsidize National Public Radio, with its upscale audience and clear political bias. We may also have to take cost into consideration in deciding issues of grand strategy, such as our position in the Western Pacific and in the Arabian Gulf. But these discretionary items are only part of the problem, and they are growing slowly, roughly in line with the economy.
The real long-term budgetary problem is in the area of entitlements, especially programs that are scheduled to grow much faster than the economy—a mathematical impossibility in the long term. Each requires sharp pencil reforms that are structural in nature.
Start with Medicaid, our health care program for the poor. It will consume roughly $4.2 trillion over the decade but will more than double in cost over that time, setting us up for real problems in the 2020s. Here the structural problem is the “matching grant” approach, in which the federal government sets minimum, but highly complex, standards and then pays half the cost. States must meet the federal standards, but may exceed them. The result of that latter provision is that state politicians may raise spending but have to raise state taxes by only 50 cents on the dollar to pay for them.
New York, which epitomizes what is wrong with Medicaid, has clever state politicians who go one better. They pass half of their share on to the counties, which must raise money from property taxes to cover it. This allows the politicians in Albany to claim credit for spending dollars of which they have to raise only 25 cents, while the Feds pick up 50 cents and the counties the other 25 cents. Guess what happened in New York? With just 6.2 percent of the nation’s population, and incomes somewhat above the national average, the state has 8.5 percent of Medicaid beneficiaries and consumes over 13 percent of all Medicaid spending. At one point a decade ago the governor’s mother was on Medicaid, along with one-fifth of New Yorkers. Is it safe to say that any program for the poor with the governor’s mother on its rolls has serious design flaws?
The problem should be fixed at the federal level: Rather than pay half the cost of whatever states decide to cover, give the states a fixed allowance called a block grant. Couple that with a sharp reduction in federal regulations so that states can design and administer programs as they see fit. This rewards innovation and thrift. It is a major principle of the Ryan budget. One can debate the amount of the block grant, but to pass any Sharp Pencil Test, block-granting Medicaid is the only way to go.
President Obama and the Democrats oppose this. One reason is that some of their special interest groups like the service-employees union are major beneficiaries of the current system. But the main reason is that Obamacare needs the matching-grant system to depress its apparent price tag. It dumps millions of people onto the Medicaid rolls, forcing states to pick up half the cost. This is a central objection of the 26 states that have sued to have Obamacare declared unconstitutional. And it is one of the gimmicks that allows Obamacare to appear relatively “low cost” at the federal level: Half the costs of an expanded Medicaid are shifted to the states.
Medicare has its own structural problem. The federal government pays 80 percent of whatever seniors spend. That includes needless tests to protect doctors from malpractice suits, end-of-life care that often goes well beyond keeping patients comfortable, and unnecessary use of scarce medical resources by beneficiaries without any incentive to economize. The Obama administration intends to curb costs by rationing care. A federal panel will prioritize procedures and deny funds for any medical procedure that doesn’t make the cut. That arrangement might work at the budgetary level, and it is used to ration care in a number of countries with socialized medicine. But the one-size-fits-all approach is crude and unappealing and brings to mind the unfortunate phrase “death panel.”
A better approach, contained in the Ryan budget and advocated by some thoughtful Democrats like Alice Rivlin, who served as director of the Office of Management and Budget and was vice chair of the Fed in the Clinton administration, is the “premium support” concept. This would provide seniors with a program similar to the prescription drug part of Medicare, and similar to the arrangement now enjoyed by members of Congress, who choose among competing health care plans that are subsidized up to a fixed amount. The plans would vary in their coverage; they might be more or less generous in their coverage of, say, end-of-life care. But the key is that the patient would choose, not a panel of federal bureaucrats. As with Medicaid, it is fair to debate the precise level of assistance, and the Rivlin and Ryan plans differ on this. But at the structural level, premium support is by far the best sharp pencil reform we can make.
Social Security, with nearly $10 trillion in spending over the next decade, is also a program that needs to be addressed. Changing benefits for those now receiving them or expecting to receive them soon is unnecessary. Drastic changes in the rules, moreover, are not needed if change is enacted now. One very progressive plan to save Social Security was put together by the Bowles-Simpson commission. Other plans, including some developed by President Bush’s Social Security Commission, would also solve the problem. What is essential is to act now, while reforms that would save the system can be introduced gradually.
Finally, taxes must be part of any broad-based solution. But the evidence is overwhelming that to continue our current highly inefficient and distorted tax system is antithetical to growth, and simply to raise rates under that system would make matters worse. Because taxes distort people’s decisions, taxpayers are worse off by $1.70 for every extra dollar of government revenue. At modestly higher tax rates than we have now, this effect would become even more severe. Taxpayers in high-tax states could be up to $4 worse off on every extra dollar the government collects.
One approach, used by both the Bowles-Simpson commission and the Ryan budget, is to broaden the base and lower the rates. Bowles-Simpson does this with rates that would raise substantial revenue, Ryan does it on a break-even basis. But the structure of both plans is the same. Broadening the base with a top rate of 25 to 27 percent would both enhance revenue and enhance the country’s growth prospects, a quintessential sharp pencil approach.
Even more dramatic would be to replace our income tax system with a system based on cash-flow. Administratively it would be far simpler to collect a single tax on business. It would also minimize avoidance based on the definition of income. As we learned in the financial crisis, “Cash is a fact, income is an opinion.”
What all of these structural changes have in common is efficiency. Ultimately that is what generates economic growth and balanced budgets. If a program that produces just a 1 percent return for the economy is funded from tax dollars taken from individuals and businesses that may get a 6 percent return on their investment, then the economy is poorer by 5 cents for each dollar spent. And if $100 million is spent on a program whose objectives could be met with an outlay of $70 million, the nation is poorer by $30 million. That is true no matter how well-intentioned the program may be, how neatly it fits into a political speech, or how much it may appease a special interest group. We must stop equating the spending of dollars with “stimulus.” Wasted dollars make us poorer, not richer. Taking a sharp pencil to the federal budget is the way both to restore growth and to lessen the burden of government debt we are leaving to our children.
Lawrence B. Lindsey served in the Reagan, George H.W. Bush, and George W. Bush White Houses and at the Federal Reserve during the Clinton administration. His most recent book is What a President Should Know . . . but Most Learn Too Late.