Over the last couple of years, there’s been a lot of speculation that traditional economic models are failing to explain the post-Great Recession world. Some are aruing that we’re in “secular stagnation”. Some think that the Fed needs to pursue more radical policy to adjust for consistently sluggish growth and stagnant wages and prices. This could be debated with regards to several areas of the economy, but I want to focus particularly on consumer spending.
Standard economic theory, (Friedman’s consumption-savings model and the subsequent buffer stock model developed by Angus Deaton and Chris Carroll) tells us that spending shouldn’t respond to the timing of income because people are always smoothing their consumption. The concept of a “hand-t0-mouth” household developed based on the idea that some households spend all of their income upon receipt and don’t save because their income just meets their consumption needs. Thus, a payday response is expected in these households for both expected and surprise income.
A recent paper uses data from a personal financial management tool to show that payday responses are actually surprisingly robust even for people with substantial liquidity, though. One possible explanation is that people feel the need to have a liquidity buffer, so even if they’re not actually constrained, they feel constrained and thus must maintain a certain level of liquidity at all times in order to satisfy their risk aversion. This is supported by other recent research which demonstrates that credit limits are highly variable. So when money comes in on payday, they can all of the sudden spend, even though in reality they could’ve spent it on the same thing with their liquid assets a day earlier.
Before I proceed, I want to acknowledge that this PFM is located in Iceland, so I’m certainly not arguing that the results are universally applicable.The data covers 2011-2015, so we also can’t say if behavior has changed in recent years relative to before the recession. I do think its opens up an interesting discussion, though. The documented behavior is so far from what models would predict for payday responses and liquidity holdings, it seems like an anomaly.
Given how attitudes have changed since the Great Recession I would speculate that people are behaving much more cautiously than in previous years. The job market is still rocky, wages are stagnant, and people don’t have a lot of trust in the financial system. This might be driving individuals to hold more liquidity than they would have 10 years ago.
If this were true, it would go a long way in explaining why the recovery has been so plodding. Like having interest rates at 0% if no one wants to borrow is compared to pushing on a string, having money in the bank if you’re too afraid to spend it is similarly useless from an economic growth perspective. Consumer spending, though it has been increasing in recent months, has shown much slower growth over the past few years years than what was typical pre-recession.
These results have clear implications for fiscal policy and how to best implement stimulus, but what do they mean for monetary policy? Well unfortunately, Central Banks are limited in their tool sets and are currently not showing any signs of attempting nontraditional policy. Knowing that people do have liquidity that they’re not spending is powerful information though, especially with the clear evidence that a 0% interest rate hasn’t stopped consumers from just letting their money sit in a bank earning nothing (or in some cases, just sitting at home). What we need to know is why. If it’s because they’re afraid of losing their buffer, this is an issue that can only be solved by restoring confidence in the current financial infrastructure or changing it into something more trustworthy so that people don’t feel the need to horde money.
It also indicates that a higher interest rate might motivate some asset shifting to more long-term savings vehicles which would reduce buffers and potentially induce more true hand-to-mouth behavior, thus making payday responses even stronger. This could potentially help boost spending and general economic activity. Or it could cause people to hold onto even more liquidity since they’d actually be making a return. If the models can’t predict what will happen then I certainly can’t either.
So we’ve got a lot of things going on:
- Low or near 0% interest rates somehow coupled with people holding onto unnecessary amounts of liquidity
- Low consumer trust/confidence in the financial system
- High variability in credit limits which leaves people uncertain about their ability to borrow in the future
Basically everything points to a lot of uncertainty, which is never good for economic growth. Combine that with low returns, and it’s a recipe for malaise.