I’m delighted to welcome Jesse Eisinger to the ranks of people who think it’s high time that we abolished — or, at the very least, significantly curtailed — the tax deductibility of interest. Paul Volcker was an early member; the CBO has been making the case for a while; and Treasury has been very explicitly in favor since February.
The last time I wrote this idea up, I quoted Dan Primack, who suggested that 65% of corporate debt interest should be tax deductible, along with 100% of interest at companies with less than $20 million in revenues. To which I added: “(And, presumably, banks, too.)”
But Jesse isn’t making that presumption: he thinks that even banks — especially banks — should be hit by such a change.
What isn’t well appreciated is how much the debt deduction helps the banks. The first way is direct: Banking is a highly leveraged industry. Banks use more debt than equity to finance their activities. The tax break makes the debt cheaper and encourages banks, at the margin, to gorge on more.
This is absolutely true. And I completely agree with Anat Admati, and many others, that banks should have less debt and more equity. But I’m not convinced that fiddling around with the deductibility of interest is the right way to go for banks. For everybody else, yes. As Treasury points out, if a company finances new investment with equity, the effective marginal tax rate on that investment is 37% — while if the investment is financed with debt, the tax rate is minus 60%. A difference of 97 percentage points.
Think about it this way: if you borrow money on your credit card, you pay off the interest with your after-tax income. That’s as it should be: no one’s seriously suggesting that you should be able to use pre-tax dollars to pay for the interest on the greetings card you bought last month. But for banks, as for all companies, the deductibility of debt makes a huge difference.
It’s worth spelling this out. To make it simple, let’s use an old-fashioned 3-6-3 banker: he takes money in at 3%, lends it out at 6%, and hits the golf course by 3 o’clock. And let’s say that the bank just acts as the intermediary between depositors and lenders.
So Fred has $1,000 on deposit, and Brenda has a $1,000 one-year loan. When that year is up, Fred’s deposit has grown to $1,030, while Brenda has repaid a total of $1,060. The difference — the profit to the bank, which the banker has to pay tax on — is $30.
Now let’s say the bank was financed 50% by debt, and 50% by equity. It still lends Brenda $1,000, but it takes just $500 from Fred, and uses its own money — prior years’ accumulated income, perhaps — to find the other $500 to lend to Brenda. At the end of the year, it has still earned $60 from Brenda, but this time it has paid Fred only $15 in interest. Which means that the bank’s profit has risen to $45.
And of course if the bank was financed wholly by equity — if it had no deposits or liabilities of any sort — then it would make a profit of the full $60 on Brenda’s loan.
So why don’t bankers use lots of equity and very little debt, if they like profit so much? Because of the power of leverage. Take that $1,000, lend it to Brenda, and you make $60. But what if you take that same $1,000 and make ten loans instead? Each loan would comprise $100 of your own money, and $900 which you’ve borrowed from Fred. On each of those loans, you take in $60 from Brenda, and pay out $27 to Fred in interest, for a total profit to yourself of $33.
And if you make $33 ten times over, that’s $330, which is a lot more attractive than a mere $60.
Right now, nothing in the tax code changes this fundamental mathematics at all. Let’s say the bank has a corporate income tax rate of 30%. Then the $60 of income on one loan becomes a post-tax income of $42, while the $330 of income on ten loans becomes a post-tax income of $231.
But what happens if you abolish the tax-deductibility of interest? Then things change dramatically. In the single-loan case, the bank doesn’t make any interest payments to depositors: Fred’s not in the picture at all. And so the pre-tax profit remains at $60, and the post-tax profit remains at $42.
But in the ten-loan case, the $330 is the difference between $600 in revenue and $270 in the cost of interest paid out to depositors. If you can’t deduct that $270 in interest, then you have to pay tax not on the $330, but rather on the $600. Which means your tax bill goes up to $180, and your post-tax income falls from $231 to $150.
Now $150 is still larger than $42. But the multiplier effect is shrinking. With tax-deductible interest, issuing ten loans got you 5.5 times the profit that you saw when you were making one loan. Without tax-deductible interest, you’re still taking ten times the risk, but your final profit is only 3.6 times what you would make by just issuing a single loan directly.
So, should we abolish or severely curtail the tax deductibility of interest even for banks? Would that be a good way of giving them a little bit of incentive, at the margin, to cut down on excessive leverage?
I’m not convinced. The way to cut down on leverage, it seems to me, is to cut down on leverage. That’s what Basel III is for, not the tax code. For banks, money is their raw material: it comes in, gets transformed, and goes out, every working day. And for any business, profit is what you’re left with after paying for your raw materials. I can absolutely get on board with making it more attractive for a widget maker to invest in its raw materials using equity rather than debt. But when debt is your raw material, I’m not sure.
Certainly any such move would make checking and savings accounts more expensive for consumers. Remember that while a bank account from the consumer’s perspective is a handy place to keep your money, from the bank’s perspective it’s a funding source — the depositor is lending money to the bank, which then turns around and lends it on to someone else. If the bank had to pay income tax on all of the interest paid to depositors, that would surely cause quite a lot of harm to the whole depository ecosystem.
That said, there are aspects of the idea that I like. It would encourage banks to make real loans to real people, at real margins, rather than engaging in clever financial shenanigans where the profit is a tiny sliver compared to the cost of funds. (Or, to put it another way, it would encourage JP Morgan (JPM) to move less money to the Chief Investment Office rocket scientists in London, and move more money to its branches for personal and small-business loans.) And in general, as I’ve said many times, our entire society needs to deleverage and move to more of an equity-based funding model.
But let’s not start by engaging the banks in a thermonuclear regulator war when we don’t really have any idea what the unintended consequences might be. There’s an enormous amount of good to be done just by abolishing or reducing the tax-deductibility of debt in the commercial sphere; so let’s begin there. If that works well, then maybe we can think about moving on to banks in some way.