By now, we've been able to observe three monetary trends developing in 2015 concerning inflation, unemployment, and the Fed's rate hikes (or the absence of them). All have shaped how the markets trade between periods of change in their respective indicating figure, and all remain inevitably interconnected. One of the most famous theoretical (and often applicable) economic correlations is between the inflation rate and the amount unemployed, Many famous economists have taken part in the debate surrounding the connection with Keynesians and Monetarists joining on either side. Big names like Milton Friedman and Paul Samuelson, both Nobel prize winners in economics, have explicitly dedicated some of their time exploring the strength and sometimes existence of this apparent correlation. In order to develop the theoretical framework behind these ideas, William Phillips, an economist from New Zealand, wrote a paper describing what is now known as the Phillip's curve, an inverse relationship between unemployment and inflation (a picture can be seen here). This became an even more controversial issue as the U.S. economy entered periods of high inflation then stagflation in two straight decades, the 1970's and the 1980's. This attracted a lot of economic thinkers to the problem which were mentioned above, and they debated complications like market expectations and long-term versus short-term correlations. Fast forward to this year, and once again, the relationship between inflation and unemployment emerges at the center of the rate hike debate.
|Data from FactSet, www.usinflationcalculator.com|
The graph above shows the movements of both U.S. unemployment and inflation rate over the past ten years on a monthly basis. The unemployment rate moved very little with the exception of the financial crisis and the recession afterward in 2008 and 2009, but inflation rate was a lot more volatile and continues to move erratically despite the emergence of relative trends. In the period leading up to the financial crisis, 2005 to the beginning of 2008, inflation became a monetary concern as housing prices were increasing rapidly and cash flow in the overall economy sped up after the dot-com bubble burst. The Fed raised rates in this period, and unemployment remained below 5% which was the apparent trade-off of higher inflation for more jobs, a confirmation of the Phillip's curve in the short run (2-3 years). After 2008, the recession reversed the pattern to low levels of inflation with high unemployment (although the loss of jobs was mainly caused by the crisis). As the Fed pursued an expansionary monetary policy, market expectations caused the Phillips curve to shift over, forcing the Fed to accept accelerated inflation for a recovery. This effect can be seen in the picture linked to the blog in an earlier paragraph. The curve shifting to the right makes unemployment gains move slower relative to inflation gains. Starting from the lowest point in July of 2009, one can observe the effect of market psychology preparing for expansionary policy as well as quantitative easing weakening the dollar substantially. This period supports the idea of an "expectations-augmented Phillips curve" that shifts when considering the expectations of the market. As the economy progresses further into the post-crisis period of expansionary policy where interests rates continue to float around 0.25% (down from 5.25% in 2008), the Phillips curve starts to become an ineffective representation of the trade-off that could happen. By 2012 and 2013, it appeared as if the Fed had been able to reach their target of 2% inflation, deemed healthy for the continuation of the recovery of unemployment and the support of the bull market rebound, but inflation began to taper off into deflationary pressures even though unemployment still slowly dropped to 6% and continued lower. This year, we find ourselves fighting negative inflation with unemployment still shrinking to 5% and possibly lower. Milton Friedman would call this an exhaustion of the Phillips curve trade-off which breaks down in the long run; instead, the relationship reverts to a point called the "non-accelerating inflation rate of unemployment" where market expectations cause the curve to transform into a vertical line. At this point theoretically, expansionary policy becomes moot, and increases in inflation will not be coupled with equally progressive employment. From 2013 to the present, the economy is seeing the correlating relationship reversed because of long-run expansionary monetary policy, as Friedman cited. Although this time, inflation and unemployment are going down together instead of increasing together like in the 1980's. There should be a widespread realization that the relation has broken down, and the Fed should respond accordingly with a change in monetary policy which will, in turn, change the market expectations with a leftward shift of the curve.
|Data from FactSet|
Adding to the theoretical framework of the Phillips curve is the New Keynesians with their ideas of aggregate demand changes (which is sometimes associated with Stiglitz). Their addition was the notion that a positive relation existed between inflation and the level of demand in the economy, and they asserted the strict short-term application of the trade-off as well. A model like this might reflect what is going on the economy now and why inflation and unemployment are moving in the ways that they are. The second chart shows the relationship between the unemployment rate and the GSCI basket of commodities which measures the performance of almost all the commodities over time. For our little study here, the commodity basket represents demand change because of price sensitivity to global scarcity (as seen in the oil glut). After years of stable prices, commodity prices drop significantly starting midway through 2014, mostly caused by a major drop in crude oil prices. Demand for commodities becomes a serious problem in the summer of 2015 as it edges on deflationary problems and causes a reduction in global trade accounts, hurting emerging markets. For this reason, consumer and producer prices became volatile and eventually cheaper as buyers pulled back. With this deceleration of demand, the global economy should see a deceleration in inflation, which has happened, but at the same time, there should be a deceleration in hiring, right? Wrong. The Phillips curve has already broken down because the expansionary monetary policy has already been exhausted. Market expectations have already changed because of the conclusion the markets have made: it is time for stricter policy. And just like we said earlier, the curve will be (theoretically) shifted to the left where the current level of unemployment is associated with less inflation. Perhaps this is where the Fed went wrong, inflation could not be stimulated to the 2% target that was set because (1) the long-turn Phillips curve had transformed to the NAIRU, and (2) the markets have already adjusted the curve because of their expectations for stricter monetary policy. That leaves this question to be asked: when should they have increased interest rates? That question is not answered easily, but it does elicit answers that should have been considered earlier. If the expansionary policy established in reaction to the crisis was for reducing unemployment, the reversal of the trade-off should have been recognized earlier. Perhaps the Fed was scared into expansionary policy for too long because of the frightening consequences of the crisis, or perhaps the Fed acted too dovish when supporting the recovery. A sudden correction August might not have happened if monetary policy had been cooled down earlier. Adjustments could have happened over time, and acute changes in inflation levels (and aggregate demand levels) would have been more gradual had market expectations of stricter policy been recognized earlier. If the December rate hike does not occur, the economy could see more corrections after more overheating off of loose policy. In the case that rate hikes do occur, we can see this relationship readjust into a healthier cyclical correlation that can reduce volatility and support growth. Then when needed, policymakers can take advantage of the short-run trade-off the Phillips curve offers during periods of decrepit economic growth.