The Problem Is Putting Weight On Made Up Categories

Megan McArdle has a piece arguing that GruberGate shows an insider problem:

So when I see journalists saying that Gruber's revelations don't matter because he's just kind of awkwardly saying something that everyone knew, I get a little jittery. I am not "everyone," and neither are any of those journalists. We're a tiny group of people with strange preoccupations who get paid to spend our time understanding and explaining this stuff. The fact that we may have mentioned it once to our readers, in the 18th paragraph, does not mean that readers read it and understood what it meant. (In fact, if you actually interact with your readers, you'll be astonished at how little they remember of what you told them, especially if you didn't go out of your way to headline it. Their minds are already crammed full of information that they need to, you know, live their lives. So they tend to take away a few big bullet points, not the piddling details.)

As I hinted at before, I think this analysis misses the real problem that Gruber's comments expose. The real problem is that distinctions between fiscal categories (spending, taxing, mandating, user fees, and so on) are totally made up and practically interchangeable. Journalists' great sin is that they report on these things like that isn't the case. They present stories as if, for instance, the word "tax" has a coherent meaning that is significant to how you should judge policies. But it doesn't.
These categories are somewhat useful, even though they are constructs, because they can sort of help you communicate what's going on. The issue is that these empty fiscal constructs have been made to correspond with normative views. A good deal of our politics involve people expressing that they are against increasing or decreasing certain fiscal categories that are, deep down, totally meaningless. And in a world where empty constructs that are often interchangeable are given massive normative weight, you're going to have severe confusion no matter what.
I don't know what any given journalist can do about this confusion. You can't, in every article, point out that a tax cut is actually indistinguishable from a spending increase, that a spending decrease is actually indistinguishable from a tax increase, and so on. It would take up too much time and would be a slog to read through.
Further, most journalists, I suspect, do not even realize that public finance categories are as empty as they are. These people don't play along with the constructs because they are useful. They play along with them because they think they are real. They think that someone who says "I am just for lower taxes and lower government spending" has actually expressed a cogent idea, when they haven't. In short, most journalists don't strike me as sophisticated insiders; rather, they seem to be just as confused about these matters as people who don't spend their time purportedly covering them.
Even journalists who are sophisticated, like McArdle, find themselves falling into this conceptual trap from time to time. Upon reading McArdle's piece, I immediately recalled an argument she purportedly made at the Wonkblog's debate about the propriety of installing a Universal Basic Income. I was not at this debate, but I recall it having been attributed to her on Twitter.
She argued, among other things, that a Negative Income Tax is better than a Universal Basic Income because it costs less. This hit a nerve in me at the time, and I wrote about how confused it was. The only way you conclude that a NIT is less expensive than a UBI is by counting the effective marginal tax rates of the NIT as spending reductions rather than taxes that are going to fund spending. But this a meaningless distinction.
To be concrete about how meaningless this distinction is, consider the following two programs.
In Program One, every adult is given $10,000 each year in cash benefits. This benefits are funded by imposing a 50% tax bracket for market income earned between $0-$20k, with the remainder of the necessary funding provided by other taxes. Under this program, a person with a $5,000 market income will receive $10,000 in benefits and pay $2,500 in taxes (for a disposable income of of $12,500). The fiscal cost of Program One is equal to $10,000 multiplied by the adult population.
In Program Two, adults are given cash benefits equal to $10,000 minus the first $20k of their market income multiplied by 50%. Under this program, a person with $5,000 of market income will receive $7,500 in benefits and pay $0 in taxes (for a disposable income of $12,500). The fiscal cost of Program Two is way lower than the fiscal cost of Program One. Program Two also involves far lower taxes than Program One.
But Program One and Program Two are the same thing. Their different effects on "taxes" and "spending" are purely a function of accounting constructs that are meaningless. It would be nonsense to say we should favor Program Two over Program One because it has lower costs and lower taxes. But that is what McArdle's purported argument for the NIT over the UBI amounts to. 
I bring this up not to make some big deal of this one bad argument, but just to use it as an example of what I take the bigger problem to be. Journalists covering fiscal policy are required to rely on constructs that are empty, but also that people, including the journalists themselves, ascribe significant normative weight to.
In a world where such confusion reigns, you're going to have this problem. In a world where we fight out what are really disagreements about who should get what by talking about our views on the size of fiscal categories that are empty and interchangeable, you're going to have this problem.