> if depreciation rules permit depreciation that is
> more rapid than actual economic depreciation, the net
> impact is a subsidy to capital.
Let's start from the top: depreciation (as opposed to say allowing the full deduction at the time it's incurred) is a _negative_ subsidy (I guess you call that a "cost"?) to capital. Reducing the depreciation period (trucks take 27 years; how old is your truck?) gets some of that back, but there's _never_ a subsidy; it might be a wash (as in the case of leasing), but it's never a subsidy.
In fact, it's usually the other way around. For instance, you may upgrade desktop computers every two years, though they are listed as five year property. So when you take them out of service (i.e., their "value" goes to zero) you write the rest of them off. It in effect shows clearly that five years is dumb.
-----
I'm not sure how we got to talking about depreciation: depreciation is a way to increase current tax revenue (by stretching out the _ultimately allowable_ deduction) at the expense of future tax revenue. AMT is also a mechanism to do this, albeit more directly by deferring a deduction that's allowable only if you have other income to get taxed on.
Depreciation is ultimately revenue neutral (in most cases!); so is AMT.
So I guess that's my point: an un-do of AMT, while complex for accountants, doesn't really impact the treasury.
>> Their framing of leasing and depreciation as Tax Evil ...
>
> it's a basic principle of tax economics. It is
> disputed by the Right, which claims any tax on capital
> is double taxation ...
I'm not sure why you bring that up: depreciation does indeed keep capital from being taxed. But it's just a time-shifter. We have this notion of a "tax year" but if all your purchases are "five year property" then your "tax five years" are the same whether you expense the cost of the computer entirely in year one (perhaps through leasing) or over five years (through ACRS). The only difference is when it happens.
> the point of the AMT is that some corps have enough
> ways to offset income to make rates irrelevant.
I'd be happy to see some of those things removed. But AMT never was a way to do that. If you pay AMT and in later years rack up _actual taxable income_ you can reduce it by AMT that you paid in the past! It's like a loan to the government with the understanding that you're covered in the future if you incur actual tax liability.
AMT started life as a political response to the shocking awareness that some 200 "rich" people weren't paying any taxes in 1968.
> the problem is that a deduction can be legitimate in principle but
> subject to excessive albeit legal use. For persons, one could limit
> total deductions, but this is just an AMT in different guise.
Why is that a "problem" ...?
The tax code is explicit on this point: there's absolutely nothing wrong with taking every single deduction you are entitled to.
But back to fuzzy math: one simple non-AMT preference item (i.e., can't count against you and force you to pay AMT if you "abuse" it) is payroll. If I have an extra million dollar "profit" (revenue minus expenses) and I decide to pay bonusses to my employees, I can (evilly, in CTJ's eyes) "avoid" taxation! Boom, that million dollars disappears, and the company pays no tax. But what really happened? The folks who got the bonus paid a lot more personal income tax. There was no tax "avoidance" (forget that personal and corporate rates aren't the same) ...
It's this kind of logic that loses me when I read CTJ's work.
Put another way: why exactly did Boeing rack up over a billion dollars of AMT charges in the past 15 years? Note that Boeing has an AMT carry-forward for that amount of time, technically an asset. So un-doing AMT will have zero long-term consequence on Boeing. It may have a nice uptick for a quarterly report, but anyone who cares about the "real" value of Boeing won't be fooled.
AMT was/is never a help, why does it hurt to make it go away?
/jordan