Feature Article: Why After a Period of High Uncertainty Persistent Deflation is More Likely than High Inflation

There are two ways the High Uncertainty regime could transition to High Inflation. The first scenario would be a substantial negative supply shock, such as a sharp decrease in the supply of oil (e.g., due to Iran making good on its longstanding threat to mine the strait of Hormuz), or massive crop failures (e.g., due to faster and larger than expected changes in global temperatures). To put this risk into perspective, just three grains -- wheat, rice, and corn -- account for 40% of all global dietary energy consumption.

Inflation caused by higher oil prices would likely quickly self-correct, as higher prices would reduce aggregate demand, which in turn would eventually lead to lower oil prices and inflation. On the other hand, a supply shock to the global food supply would very likely take longer to correct.

In the second scenario, the US Government would increase spending to fight a potentially deep downturn, finance this spending with debt, and then monetize that debt by having it purchased by their central banks. A slightly different version of this scenario has been offered by Bridgewater’s Ray Dalio. He posits that entitlement payments will rise faster than tax collections, and widen the US Government deficit at the same time as investors (both domestic and foreign) are increasingly unwilling to purchase it unless yields significantly increase. Fearing that this will choke off economic growth, the US Government successfully pressures the Federal Reserve to purchase more government debt, which holds down yields but at the cost of a sharp increase in the money supply, which eventually precipitates a run on the dollar and a rise in domestic US inflation.

Our assessment of this second high inflation scenario will begin with this familiar macroeconomic equation: MV=PQ. On the left side, M represents the money supply (i.e., a stock) and V (for velocity) represents the rate at which a given stock of money is spent (i.e., a flow). On the right side, P represents the price level (i.e., an increase representing inflation and a decrease representing deflation), while Q represents the real output of goods and services in an economy over a give period of time (i.e., real GDP). Hence, P x Q equals nominal Gross Domestic Product.

In theory, if monetary velocity (V) and real output (Q) remain constant, an increase in the money supply (M) should produce an increase in the price level (P). Yet despite a very substantial increase in the money supply after the 2008 financial crisis, we did not see a sharp increase in inflation. Why was that?

Let’s look at how the individual components of the MV = PQ equation behaved between December 2008 and December 2017. The cumulative increase in the US price level (as measured by CPI) was 17%. The cumulative increase in real output was 19%. The cumulative increase in nominal GDP was 39% (since the PQ relationship is multiplicative, not additive).

But the M1 money supply increased by 124% over this period. The reason the US did not experience a sharp increase in inflation despite an unprecedented increase in the money supply (via the Fed’s Quantitative Easing policy) was due to an equally unprecedented fall in monetary velocity.

The obvious question is why this occurred. Why did the private sector choose to save money rather than spend it? In a 2014 blog post (“What does Money Velocity Tell Us About Low Inflation in the U.S.?”) economists at the St. Louis Federal Reserve Bank offered to possible explanations: “A gloomy economic outlook after the financial crisis” and “the dramatic decrease in interest rates [to zero or negative real yields] caused a portfolio shift away from interest bearing assets and into cash.” In essence, the latter describes a liquidity trap (see our May 2001 article, “What’s a “Liquidity Trap” and Why Should I Worry About It?”).

Let’s now consider how these results could differ if in the future the federal government increases deficit spending and finances it via the issuance of bonds that are bought by the Federal Reserve, because of a lack of market buyers at what the government perceives as an acceptable yield. Let’s further assume that this produces an increase in both real output (Q) and the money supply. Once again, the question of whether this results in a sharp increase in inflation algebraically turns on the likely behavior of velocity.

Our view is that if the increase in government deficit spending takes place in the context of weakening private sector demand, with the current level of private sector debt overhang, and real rates still low or negative, it is hard to see why velocity would increase from its present low level. But that said, at the current level, there is much less room for it to further decline, so arguably that could produce some increase in inflation – but probably not that much unless the monetary expansion and deficit spending were truly massive. But in the current political environment, with the Democrats poised to control the US House of Representatives, it is hard to see how that could come to pass.

Moreover, there are also three other important factors to consider. The first is the assumption that central bank monetization of part of the US deficit will be necessitated by a lack of other buyers for US government debt at a yield deemed acceptable by the Treasury. In a weak economy, that may also be facing a Eurozone crisis and a more aggressive China, this assumption may not hold true. Even if the price of gold skyrockets, at some point it cannot absorb the likely savings flows seeking a low risk haven in the storm. And while not as attractive an investment as in the past, US Treasuries may still be the least ugly of the alternatives on offer (e.g., because of the superior breadth and liquidity of the US Treasury market).

Second, it seems unlikely that the US Federal Reserve will give up its independence (and the anchor on inflation expectations it provides) without a political fight. Any assumption that an administration can easily win that fight and win approval for a large increase in deficit spending is likely to be wrong.

Finally, given the amount of supply capacity that has built up in the world economy relative to what has been weak demand in recent years (a powerful if latent deflationary force, as we have noted), it may also be the case that the potential increase in prices caused by rise in demand driven by government spending will be offset by a sharp increase in supply (unless, of course, that supply increase is blocked by much higher trade barriers or other political actions).

To be sure, history – and the present – warn us that hyperinflations can occur. But in the case of the United States, there are also a series of countervailing factors that seem likely to offset most (but not all) of the potential causal pathways to the High Inflation Regime.

If you have any questions about anything we have written in this issue, please don’t hesitate to get in touch, at contact@indexinvestor.com