Trust me, I wouldn’t waste your time with just any new economics paper. But economist Olivier Blanchard just released an analysis that is so germane to issues of great importance to economies across the globe that attention must be paid. Besides being one of the world’s top macroeconomists, I’ve long admired Blanchard’s work for its relevance to current policy and for his careful effort to keep his political thumbs off the scale, a critical asset in these partisan times.
His paper, “Public Debt and Low Interest Rates” (hey, I said he was smart, not catchy), injects some extremely important facts into discussions about fiscal policy that have long been characterized by assertions, biases and fearmongering. If we learn and apply the lessons I take from this paper, we can stop making mistakes that have been terribly costly to the well-being of millions of people.
In my own interpretation — as noted, Blanchard’s careful not to push too hard on how his work should plug in to current fiscal debates — his paper argues that we’ve been focusing on the wrong thing. Given the actual and expected levels of the key economic variables he scrutinizes — interest rates and growth rates — we should not be overly worried about deficits and debt. We should certainly not reduce them when they are necessary to support weak economies, as “austerians” have done, especially in Europe but also here, in the expansion’s early years.
However, deficits still matter, they still invoke risks, and it matters a great deal what the public sector spends its borrowed money on. Offsetting recessions, investing in public goods (human and physical capital), job opportunities for those left behind, providing retirement and health security for those who lack it, sure. But significantly raising the debt by providing tax cuts to those who don’t need them unnecessarily invokes risks, even if their probability is lower than we thought.
How does his work suggest these conclusions? The key points are disarmingly simple, and they’re ones I have written about before in this column. Part one is this: When a country’s growth rate is higher than the interest rate on its debt, the fiscal costs of sustaining its debt levels are somewhere between zero and low. The reason is that even if the government does not raise taxes to offset its higher debt, the ratio of debt to gross domestic product will decrease rather than explode over time. Part two: For most of the period covered by Blanchard’s research (1950-now in the United States), g>r, i.e., the GDP growth rate has exceeded the interest rate (same with the U.K., the euro area and Japan).
For example, after making some key adjustments to the relevant interest rate on public debt, Blanchard writes: “Over the period [since 1950], the average adjusted rate has been substantially lower than the average nominal growth rate, 3.8 percent versus 6.3 percent.” I should note that pathbreaking work by my colleagues Richard Kogan et al found a similar relationship in U.S. fiscal data back to 1792.
There’s one weedy detail, however, that Blanchard brings to the table that tilts slightly in the other direction. One reason the extent of public debt matters is because it tends to lead to less private capital — machines, structures — than would otherwise accumulate. Whether that’s a problem has to do with a slightly different comparison: that between the return on capital — what it provides for us to consume and invest — and the growth rate. In some of Blanchard’s analysis, that comparison shows the return rate to be a bit higher than the growth rate. This implies that, even if debt has little or no fiscal cost, it may impose a cost on society in the form of lower output and consumption.
Here again, however, the correct interpretation (mine, not his), is that because, under certain conditions (e.g., a period with high returns on private capital investments), debt incurs a cost to our future welfare, we need to be mindful of what we’re spending it on.
To better understand that key point, I need to underscore something Blanchard leaves out of his analysis (not a critique — this is outside the goal of his paper, which was to nail down the points above). In this analysis, all public debt is created equal. At first blush, anyway, the research doesn’t distinguish between what I’m going to call GD and BD, or good debt and bad debt (to the extent the impacts of GD and BD flow through differently to growth and interest rates, they do show up in the paper).
The distinction between those two poles is hugely important. No matter how low interest rates are, it will always make more sense to borrow for GD than BD. The challenge, of course, is that we need a definition of GD that works for most of us. Mine is simple: GD invests in people and places that need the help; BD does not.
Thus, a countercyclical Keynesian stimulus, meaning deficit spending in a recession to offset a demand contraction leading to higher unemployment, is GD, because under those conditions, a lot of people need help. However, what I call “upside-down Keynesianism” — stimulating an economy that’s already closing in on full employment with tax cuts to the wealthy and corporations … well, that’s some seriously BD. Instead, had the $2 trillion in deficit-financed tax cuts instead gone to poverty reduction, jobs for those left behind, housing for those lacking shelter, affordable health and child care, productive infrastructure investments the private sector won’t make … well, now we’re talking about GD.
Finally, one of my greatest hopes for this paper is that Blanchard’s straightforward analysis, in tandem with his stature, puts a knife through the heart of austerity economics, the heedless, reckless, premature removal of fiscal support from weak economies for no good reason.
If that occurs, we will be witnessing something all too rare in economics: a bit of sensible analysis that led to a change in policy that prevents a lot of people from being made worse off. Sounds simple, but replacing bad, ill-founded ideas with good, analytically sounds ones is way harder than it should be, and it’s not getting any easier.
Here’s to hoping this paper helps change that.