The current inflation surge continues to attract attention. Fed Chair Jerome Powell insists that the surge is ‘transitory’, emanating pandemic related supply shocks that will dissipate as world production comes back online. Others, I included, worry that the current extreme accommodative monetary stance along with massive fiscal expansion will not resist these prices increases creeping into inflation expectations, leading to a persistent increase in inflation.
The G7 has luxuriated in 30 years of lower, stable, and declining inflation rates. The combination of demographics – aging populations – creating a large savings glut and inflation-targeting central banks accommodated this stability even through several crises. Moreover, this credibility owes much to the painful adjustment that former Fed Chairman Paul Volcker implemented in the late 1970s and early 1980s. The double-dip recession of 1981-1983 persistently reduced inflation expectations with monitoring by central banks.
Clearly, the world’s central banks, especially the Fed, want to avoid what they faced in the second half of the 1970s. But it seems, in some quarters, they have forgotten how we got there. Some of this amnesia is generational: not one G3-focused market participant under the age of 50 experienced even the mild inflation uptick in the early 1990s that stemmed from a monetization of the Thrift industry debacle. Those of us in emerging markets, especially in Latin America, have fresher memories of inflation. But even there, only Argentina and Venezuela cannot seem to lower inflation to civilized levels.
Credibility is key to what happens to inflation. The way to maintain credibility is to ensure any deviation from normal policy is temporary and mild. So far, most arguments simply assume the current aggregate price surge will prove ‘transitory’, as in the most recent past. But I have never witnessed a monetary expansion like this in the US, although I have many experiences with such policies in Argentina, Brazil, Chile, Mexico, Peru, and Venezuela, especially during the debt crisis of the 1980s.
If monetary policy stays too loose for too long, this credibility will start to erode – like in the late 1960s and early 1970s, which ultimately led to Volcker’s so-called draconian treatment. If central bank policy is so loose that they have too far to tighten when the time comes, we will reproduce something like the inflationary 1970s. It is much easier to bail an inch of water out of a boat than a foot; the race to avoid sinking becomes more frenetic and difficult. We know one thing: central bank credibility is transitory if abused.
Moreover, how do we know we are in an inflationary environment? Some concentrate on the supply side, others the demand side. But in an inflationary environment, the difference is hard to tell. Remember the formation of OPEC in 1973-4 and the huge increase in oil prices it produced. Many shrugged off the price increases as a one-time event – much like the current pandemic-induced price increases – a supply-side shock that will correct itself. So, no need to tighten money or stop producing it. This then led to increases in prices upstream with firms, which were passed on to final prices and then back to wage demands.
Here, we decompose the recent shock into aggregate demand (AD) and aggregate supply (AS) shocks for the US and then Brazil. Not only will this offer insight into the genus of the shock but also how much adjustment is necessary.
Decomposing Aggregate Demand and Aggregate Supply Shocks
We decompose into AS and AD shocks à la Blanchard and Quah (1989) and Quah and Vahey (1995). The idea is that supply shocks affect both output (positively) and inflation (negatively) permanently, while demand shocks affect only inflation in the long run and have no bearing on output. The technique uses a VAR as follows:
- Inflation=coefficients* past values of inflation and real growth + Coef1*Demand Shock + Coef2*Supply Shock
- Real growth=coefficients* past values of inflation and real growth + Coef3*Demand Shock + Coef4*Supply Shock
The residuals of these two equations then are:
- Resid 1 = Coef1*Demand Shock + Coef2*Supply Shock
- Resid 2 = Coef3*Demand Shock + Coef4*Supply Shock
This gives us two equations in the two unknown shocks. The estimation of this is called structural factorization. We can invert the relationship to back out the two shocks. But beforehand, we assume the sum total of demand shocks does not affect real GDP growth. Accordingly, we can set Coef3 to 0. That makes it relatively easy to invert the two equations to reveal the shock (the structural residuals). We use logarithmic growth and inflation rates in the regressions.
AD and AS Shocks in the US
Performing this decomposition for US GDP yields the representation of AS and AD shocks (Chart 1, 12-month rolling sums). For most of the last 20 years, negative supply shocks have dominated, especially the largest during the 2008-9 crisis and the current COVID-19 epidemic. These cumulative supply shocks had improved before the pandemic, starting in 2014, but still reflect the lack of long-term productivity growth. Economic stimulus policies have buoyed the demand side, rarely dipping into negative territory.
The Fed’s quick action in 2008-9 avoided a complete financial meltdown. But the banking system was in much better shape entering COVID-19, with the leverage of individuals having shrunk by a record proportion. I believe the slow recovery after 2008-9 had more to do with continued deleveraging and financial system convalescence than any perceived lack of stimulus. Certainly, the positive demand shock starting in 2007 belies the narrative of insufficient stimulus. This alternative view is also consistent with the slow supply-side response in 2008-9 compared with the current recovery, which is a quick and strong snapback.
Remember, the numbers in Chart 1 go only to March 2021. We shall see more when the 2Q 2021 numbers come out. If not met with a counter shock, positive demand shocks are inflationary, as are negative supply shocks. Both indicate higher inflation. How permanent depends on the evolution of the shocks. But we know there are significant positive demand-side shocks in the pipeline with nothing pushing back other than the credibility of the Fed and other central banks. This credibility is significant and important. But as we saw in the 1960s and 1970s, once credibility starts to wane, restoring it is hard. I will spare you the bromides about toothpaste and horses, but I think my point is clear.
AD and AS Shock in Brazil
We can compare the decomposition with Brazil (Chart 2). Its central bank has increased its inflation-fighting credibility since adopting inflation targeting in the 1990s, but still is challenged. As in the US, supply shocks have dominated. But unlike the US, Brazil has suffered several negative demand shocks since 2016, mainly after the government implemented fiscal adjustment and rules to curb inflation and debt growth. The Brazilian government has undertaken more fiscal support than most emerging markets. Moreover, the supply-side recovery is a snapback from a deeper drop than in 2008-9.
Analogies for the Current Recovery
The 2008-9 analogy that many US analysts used was not appropriate for the US. But Brazil’s experience then was. They suffered a huge supply shock. The government started stimulating through massive spending and credit expansion after the supply-side recovery was well underway. The excessive stimulus led to unsustainable growth rates and inflation, eroding the credibility of macroeconomic policy, and it was followed by poor, stagflation performance. I fear we will see the same in the US.