If a budget deal is reached, officials will undoubtedly boast about how much their deal “reduces the deficit.” They will use numbers like $1 trillion, $2 trillion, or even $4 trillion of deficit reduction*.* You should be wary of such numbers, which are easily gimmicked.

In theory it seems easy to calculate such a number:

- If there is no deal, over the next 10 years total deficits will be X.
- With a deal, over the same timeframe total deficits will be Y.
- Therefore, the deal will reduce deficits by X-Y.

The trick is that X is not well defined, and so X-Y is suspect.

- Does X assume that troops in Iraq and Afghanistan will be drawn down rapidly as the President has decided, or instead continue at current levels? The deficit difference over ten years is about $1.1 trillion.
- Does X assume the “Bush tax cuts” will expire at the end of 2012? Does it assume that Congress will, once again, change the law so the Alternative Minimum Tax does not suddenly bite millions more taxpayers, as it would under current law? The deficit difference of both policies combined is about $3.8 trillion.
- Does X assume that Medicare payments per service to doctors will suddenly be cut, as they will under current law a few years from now? The deficit difference over ten years is about $250 billion.

CBO says budget deficits *under current law* will total $7 trillion over the next ten years. But by making certain assumptions about how the realistic future (without a deal) will differ from current law, one can redefine the starting point for measurement, X, to be as high as $12 trillion over that same timeframe. For any given resulting level of deficits Y, the bigger your starting point X is, the bigger your claimed deficit reduction X-Y appears to be.

It’s pretty clear the White House is doing this aggressively at the moment, as they try to make the deficit reduction effects of the policy changes being discussed look large.

Suppose, for instance, that a budget deal would result in $5 trillion of total deficits over the next decade (Y = $5 trillion). If we compare that to *current law*, then the deal will “reduce future deficits by $**2** trillion.” If, however, we compare it to a starting point in which taxes don’t increase, troops aren’t reduced, and Medicare payments to doctors aren’t cut, then that same deal, those same policies, and that same $5 trillion of deficits will “reduce future deficits by $**7** trillion.” By changing the baseline, **we can make the same deficit reduction package look $5 trillion bigger**.

It is far better to evaluate a deal by looking only at Y, the deficits that would result from the deal, rather than at X-Y, the change in those deficits from an arbitrarily defined starting point. In the first example, instead of asking “Is $2 or $7 trillion of deficit reduction the right amount,” we should ask “Is a policy resulting in $5 trillion of deficits over the next decade an acceptable outcome, or do we need to do more?”

Looking only at Y is better for two reasons. It’s harder to gimmick Y than to gimmick X. Also, Y measures the results, which is what we *should* care about. Washington instead wants to measure itself for changes they propose relative to the status quo, even while they disagree on what the status quo is.

Policymakers will say “Our deal (or proposal) reduces the deficit by

Let’s be a little more precise. We will be looking at deficits over a fairly long time frame (10 years), and we are talking about huge dollar amounts. The best way to evaluate the deficit impact of a budget deal, if one occurs, is to look at **the budget deficits that would result over the next decade, measured as a share of the economy**. Then evaluate that result with a three-pronged test:

- How do the resulting deficits compare to the historic average of 2% of GDP?
- How do the resulting deficits compare to the level you’d need to hold debt/GDP constant, about 3% of GDP?
- What does the path of those resulting deficits look like over the next decade? Are deficits growing, flat, or shrinking over time?

We should care not just about deficits, but also about levels of taxes and spending. I will write about that soon.

(photo credit: Leslie Duss)