May 13 (Bloomberg) -- My colleagues Ezra Klein and Evan Soltas wrote about the Internal Revenue Service's Tea Party scandal this morning:

"The IRS does need some kind of test that helps them weed out political organizations attempting to register as tax-exempt 501(c)4 social welfare groups. But that test has to be studiously, unquestionably neutral."

That’s a nice thought. But is such a test even possible? The IRS didn’t make this mess because its employees are stupid or because they have a political vendetta. It’s because they’ve been given an impossible task: figure out which organizations have missions that are “primarily political” -- and come up with definitions for “primarily” and “political” that are neither vague nor politically charged.

Note, when we talk about these groups as “tax-exempt” all that means is they don’t pay tax on their own income. A 501(c)4 group can’t accept tax-deductible donations because it’s not a charity (those are 501(c)3 groups). Instead, a 501(c)4 is a “social welfare” organization: That is, it is supposed to produce benefits that are broadly enjoyed, rather than producing private profits for its funders. By law, (c)4 groups are free to lobby without limitation and also may engage in electoral activities so long as that is not their primary purpose. The main reason political donors want to channel their funds through (c)4s rather than other independent expenditure vehicles isn’t a tax advantage; it’s that (c)4 organizations are not required to disclose their donors.

Since 2010, when the Supreme Court's Citizens United decision led to an explosion in applications to form such groups, the IRS has tried to thread Ezra’s needle and failed completely. Their first idea was to look for key words like “Tea Party” in organization names. Realizing this would show unfair political bias, they switched to focusing on statements about orientation (“political action type organizations involved in limiting/expanding government”).

Realizing these approaches were both subject to accusations of political bias, in 2012 the IRS traded them for one that meets Ezra’s neutrality test: The IRS will closely examine “organizations with indicators of significant amounts of political campaign intervention.” But this standard is too vague to be useful.

Who will fail this test? Carolyn Duornio, a Pittsburgh-based partner in Reed Smith’s tax-exempt organizations practice, notes that some people do apply for (c)4 status and say upfront that their intended purpose is electioneering. Those applications can be rejected.

But in the political gray areas where many c(4) organizations operate, seeking to promote an ideology rather than the election of specific candidates, she argues that the IRS has no good choice but to approve applications and examine behavior after the fact.

“If what they’re saying is that they’re promoting ideas, and they are promoting ideas, then you have to let it go,” says Duornio.

I asked Duornio if there’s a good way to separate (c)4 organizations that are too political from those that are not. Her answer: “Not really.”

Gail Harmon, partner at Harmon, Curran, Spielberg + Eisenberg, is similarly pessimistic about the IRS’s ability to do good (c)4 gatekeeping. “Many of the standards that they could come up with would be subject to challenge because of their vagueness.”

Harmon has a couple of suggestions for the IRS, but they might not go very far. She suggests asking organizations seeking (c)4 status whether they expect to make expenditures that would be reportable to the Federal Election Commission or to state election officials. But she also notes that organizations applying for tax-exempt status could simply say they don’t know.

And maybe the IRS needs to train its staff better. Harmon argues that the IRS has lost experienced staff, including a lot of long-serving “nerds” with extensive expertise on non-profit compliance. A better trained staff with better supervision from Washington might handle these issues more gracefully. But they’d still be tasked with applying vague standards to determine excessive political activity.

The best hope might be to mimic the IRS’s more effective supervision of (c)3 organizations. But there are three reasons to think that might not work -- beyond the fact that the IRS gets to draw on decades of experience and case law in watching over charities and is only now beginning to grapple with political (c)4 groups.

One, a (c)3 organization isn’t allowed to engage in electioneering, so the only challenge in policing them is figuring out what electioneering is. With a (c)4 group, you have the added challenge of determining what the group is “primarily” up to.

Two, losing a (c)3 designation is more devastating than losing a (c)4 designation; the former comes with huge tax advantages while the latter’s main value is in preserving donor anonymity. So, (c)3 groups are less likely to push the envelope.

Three, think tanks, which are incorporated as (c)3 organizations, have in fact been getting more political over the last decade. They’re not generally running television ads (although the Heritage Foundation has a c(4) arm that does so) but their activities are more and more clearly intended to directly influence the political process. This slippage happens because, as with (c)4 groups, deciding what counts as “political” is subjective and therefore difficult to enforce.

So what should the IRS do about these groups and their proliferation of untraceable political money? My inclination is to say there is no good way to stop (c)4 political activity and instead Congress should seek to require more disclosure of donor information. If a (c)4 had to disclose its donors, it wouldn’t be particularly advantageous over other vehicles for independent political spending.

On disclosure, Congress must operate within a 1958 court decision overturning an Alabama law that was designed to intimidate the NAACP into ceasing operations in Alabama by forcing disclosure of its members. But this isn’t much of a barrier. More recent court decisions make clear that this requirement doesn’t apply to organizations directly influencing elections that don’t have reason to fear intimidation of their donors. Rick Hasen, a professor of election law at the University of California-Irvine who writes the Election Law Blog, e-mails that “the problem with disclosure on the federal level is political, not legal/constitutional.”

A better disclosure requirement might not be satisfying to those who hope to stop political activity through (c)4 entities, or more broadly to discourage independent political expenditures. But it’s a much more realistic option to discourage (c)4 abuse than somehow inventing a non-politicized IRS review process.

(Josh Barro is lead writer for the Ticker. Follow him on Twitter.)