Trying to make sense of Donald Trump's comments can be a risky business, but it is actually possible that he got it right and the NYT's Neil Irwin got it wrong on dealing with the national debt. Irwin had a NYT Upshot piece in which he discussed Trump's comments on monetary policy. Trump essentially endorsed the low interest rate policy being pursued by Janet Yellen and indicated that he would seek to appoint a Fed chair who would continue to follow this policy. (FWIW, we have yet to hear anything from Secretary Clinton on what sort of people she might appoint to the Fed.)
Irwin goes through Trump's logic on low interest rates and agrees that it is essentially right. But then he turns to Trump's comments on buying back debt at a discount:
"But Mr. Trump also suggested something that would represent a radical shift in United States policy if we take him seriously. 'I’ve borrowed knowing that you can pay back with discounts,' he said. He added, 'Now we’re in a different situation with the country, but I would borrow knowing that if the economy crashed, you could make a deal.'"
If Trump is suggesting that we will get bondholders to take a haircut on debt because the government would otherwise go bankrupt (something Trump has repeatedly done with businesses he owns) then Irwin is right. This would be a radical departure and is frankly almost inconceivable (we could just print the money).
But there is a way this can make sense. If interest rates rise, the situation Trump described, the market value of long-term debt falls. For example, a 30-year bond issued in 2015 at an interest rate of less than 3.0 percent, might sell for less than 70 percent of its nominal value if the long-term interest rate crosses 6 or 7 percent, which it certainly could.
The Treasury could buy up long-term debt in the market at its current market value, and replace it with new debt that paid a higher interest rate. This won't change the interest payments owed by the government, but it would reduce the nominal value of the debt outstanding. There is no logical reason to take this sort of step, except that many of the Very Serious People in Washington assign great value to the ratio of debt to GDP. In fact, before the discovery of the famous Excel Spreadsheet Error, many people in Washington thought something really bad happens to the economy if the debt to GDP ratio exceeds 90 percent. (It is worth noting that the ratio of interest payments to GDP is near a post-war low.)
In a context where fiscal policy debates among the country's leading experts are guided by superstition, it can make sense to do an empty gesture to appease the powerful. This would be like agreeing to delay construction on a major infrastructure project, so that it didn't begin on Friday the 13th. The delay of course makes no sense, but if powerful members of Congress and leading policy experts think that bad things happen on Friday the 13th, then the delay might make sense as a way to let the project move forward.
Trump may in fact have been proposing that the United States threaten a default on its debt, as Irwin argues, but it is also possible that he was suggesting a strategy to humor superstitious politicians and budget wonks. The latter would be a very good policy.
Note: This is corrected from an earlier version which had the direction of bond price movement reversed. Thanks to those of you who called this to my attention.