Prelude: the 1960s: At the beginning of the 1960s, inflation in the US was low - below 2% (on both GDP deflator and CPI) until 1965. By the end of the decade, inflation had more than doubled to around 5%. That is, the foundations for the double-digit inflation rates of the 1970s and early 1980s had already been laid. What was behind this rise in inflation? Essentially, a shift in Fed - and ultimately societal - preferences towards high employment at the expense of low inflation. Higher inflation was the result of competing claims on society's resources in the conflict generated by the need to finance the Vietnam war while keeping up social spending. Importantly, the international monetary arrangement of the time, the Bretton Woods system, ensured that other countries had to import the Fed's monetary stance by buying up the glut of dollars. Through this channel, the shift in Fed preferences ultimately led to the demise of Bretton Woods, as ‘peripheral' economies with different preferences (Germany, France) were reluctant to follow expansionary US policies.
Incidentally, there is a more general point here: societies' attitudes towards inflation are not immutable - and central banks in democratic societies are attuned to this even if they are institutionally isolated to withstand short-term political pressures. In times of plenty - when growth in productivity and living standards is fast - nobody needs inflation. It is in the lean years that the pressure on policy-makers increases to ‘do more'. This ‘more' is usually in excess of what is compatible with low and stable inflation.
Crescendo: the early 1970s: President Nixon's closing of the gold window in August 1971 meant the beginning of the end of the Bretton Woods system of fixed exchange rates. It thus removed the nominal anchor on which post-war Fed policy had been based. Fed Chairman Arthur Burns' desire to facilitate President Nixon's re-election in 1972 further contributed to a very expansionary monetary policy stance (fiscal policy was also very expansionary prior to the same election). Overall, during 1971 and 1972, global liquidity increased substantially: ‘global' narrow money supply growth surged into the double-digits in 1971 and 1972. With this came rising inflation. In response, the Fed tightened policy in early 1973 and the economy began turning down shortly after (money supply M1 peaked at 8.4%Y in 1Q73 and, on some metrics, the real economy also peaked in early 1973).
But what about oil? Readers familiar with the conventional wisdom will have missed references to crude oil prices in the above story. First off, note that the price of oil (and of commodities more generally) will, of course, depend on the state of the economy: faster actual and expected output growth will drive up demand and prices; and low real interest rates will reduce supply and thus increase prices (the owner of an oil well can extract the oil, sell it at the current spot price and invest the proceeds at the going real interest rate - the lower this real interest rate, the lower the foregone interest of not extracting). The global economic boom facilitated by expansionary monetary policy did put upward pressure on oil prices. However, due to the institutional peculiarities of the market back then, oil production kept increasing without the price going up until global capacity was virtually exhausted. It was this fact that allowed OPEC to unilaterally declare a higher price in October 1973. In short, the jump in oil quotes was due to the price rising (and quantities falling) to the market clearing level. Note that the oil price hike occurred well after the first signs of a downturn in the real economy.
Finale: 1970s continued and early 1980s: With the economy plunging into a recession, the Fed proceeded to stimulate the economy again. With the benefit of hindsight, there are good reasons to believe the Fed overstimulated following the 1973-75 recession because it overestimated the economy's growth potential (this had decreased because of the productivity slowdown) and thus the amount of spare capacity in the economy. Other central banks followed the Fed, which created a worldwide boom, negative real interest rates and rising inflation. Around 1978, the Fed turned restrictive once more (M1 growth peaked at 8.5%Y in 3Q78) and the economy experienced a brief, but sharp recession in 1Q-2Q80. Inflation eventually peaked in the double-digits and it took another Fed-induced recession under Fed Chairman Volcker in 1981-82 to finally defeat inflation.
II. Back to the Future
How close are we to a 1970s scenario? There are similarities as well as differences.
Similarities
• An international monetary system with ‘peripheral' economies importing the Fed's monetary policy stance through a currency peg or quasi-peg (see "QE20", The Global Monetary Analyst, October 13, 2010). It is, of course, no coincidence that the current monetary arrangement has been dubbed ‘Bretton Woods 2'.
• A glut of dollars due to superexpansionary Fed policy: China, Germany and others have made little effort to hide their discomfort over QE2, and the dollar fell substantially in foreign exchange markets before a renewed flare-up of the euro area sovereign crisis allowed it to recover somewhat.
• A two-track global economy: Similar to the early 1970s, the rest of the global economy is doing just fine - indeed many of the EM economies of the ‘Bretton Woods 2' periphery are, if anything, in danger of overheating. Note that by 2012, EM will make up 51% of the global economy under our coverage (measured in PPP weights) and the G10 the remaining 49%. And we know from recent experience that red-hot EM economies have a voracious appetite for commodities - for example, 33% of the increase in global oil consumption between 2003 and 2006 came from China.
• Structural change which lowers the economy's speed limit: The crisis has likely destroyed potential output and reduced the potential growth rate over the medium term as consumers deleverage and many economies retool from shrinking sectors such as construction towards expanding sectors such as manufacturing. Thus, again, central banks may be overestimating the amount of spare capacity in the economy.
Differences
• Structure of the oil market: Unlike in the early 1970s when the price of oil was fixed, oil quotes now adjust instantaneously to changes in the economic outlook. Indeed, since Fed Chairman Bernanke hinted at QE2 in August, inflation expectations have risen and oil has rallied. (Incidentally, that is as it should be: like stocks, oil is a real asset which should provide inflation protection in the long run.)
• The US economy is weak and unemployment is high: Yet US data have been improving before the Fed bought even a single QE2 dollar's worth of securities. Just as importantly, what matters for inflationary pressure is not the absolute level of the unemployment rate but the distance from the speed limit (the level of unemployment consistent with stable inflation). As mentioned above, it is very likely that, in the post-crisis world, this speed limit is substantially lower (the unemployment rate consistent with stable inflation substantially higher): inflationary pressure could therefore emerge at higher unemployment rates than before the crisis.
• The structure of the economy: no wage indexation: There is no question that widespread wage indexation contributed to the persistence of high inflation in the 1970s. However, the lack of wage indexation nowadays does not mean that inflation cannot rise in the first place. Besides, wage indexation was put in place in response to high inflation. That is, just because there is no wage indexation now does not mean that higher inflation in the future would not result in the indexing of wages.
All of this does not, of course, make inflation or stagflation a foregone conclusion. Yet in a global economic environment with the following factors, we think that in the medium term the risks are skewed towards an overheating of the global economy (especially in EM), elevated commodity quotes and ultimately higher inflation:
• The Fed will want to err on the side of caution with respect to draining excess reserves (a premature tightening of reserve requirements may have led to a recession in 1937 as banks responded by slashing lending in order to restore their desired level of excess reserves - see "Reversing Excessive Excess Reserves", The Global Monetary Analyst, October 28, 2009);
• The Fed will want to err on the side of caution with respect to normalising policy rates (Japan's premature exit from ZIRP (zero interest rate policy) in 2000 will likely serve as a deterrent) - this will likely keep real interest rates low;
• It is difficult to estimate the economy's speed limit (indeed, easy to overestimate the degree of slack in the economy);
• More generally, risk-averse global central banks prefer the threat of inflation to the one of deflation (see "Better the Devil You Know", The Global Monetary Analyst, August 18, 2010); and
• Public and private debt levels are very high, creating incentives to generate or acquiesce to inflation (see "Debtflation Temptation", The Global Monetary Analyst, March 31, 2010).
SARB Cuts Repo Rate 50bp on Improved Inflation Prospects
The South African Reserve Bank (SARB)'s Monetary Policy Committee (MPC) reduced the country's policy repo rate by a further 50bp to 5.5% on Thursday. This decision was largely supported by an improved outlook for domestic inflation, as well as concerns about the anaemic pace of the global - and as a corollary, domestic - recovery. The decision was also in line with our qualitative analysis, and the decisive signal generated by our proprietary SARB EazyMeter for this meeting (90% probability of a rate cut).
The MPC now expects CPI inflation to average 4.3% in 2011, before rising marginally to 4.8% over 2012. These latest forecasts are lower than the previous indication of 4.8% and some 5.1%, respectively. While the SARB's 2011 forecast is exactly in line with ours, its 2012 profile is much lower than our forecast of 5.5%. According to the MPC statement, the improved forecasts were based on the combination of downward revisions to administered price expectations, a stronger exchange rate than previously expected, as well as recent lower-than-expected inflation outcomes which have contributed to a lowering of the starting point of the inflation forecast.
Strong Currency and Lower Admin Price Outlook Help Improve Inflation Profile
First, we believe that the SARB has now lowered its expectations of electricity tariff increases: In its October 2010 Monetary Policy Review, it was careful to mention that, at the September MPC meeting, it had made provisions for electricity tariff increases of 20 % per annum for the next three years.
At that stage, it probably expected further significant increases in electricity tariffs from municipalities in the August CPI data that were to be published at the end of the month. Given that the August CPI data only showed a 1.5%M increase in tariffs, thereby taking the cumulative increase between July and August to 17.8%, the SARB may have taken the view that future increases are likely to come in below 20% as well. Also, the SARB may have taken a more benign view on other administered costs such as education, health and communications, given that increases in these costs are sometimes backward-looking with regard to CPI, and appear to be decelerating.
Second, is the ZAR. The MPC also provided reasons to believe that the rand could trade stronger for longer than it had previously expected. Most importantly, it mentioned that the recent resumption in quantitative easing by the US indicates that monetary policy in that country is likely to remain highly expansionary for some time. The search for yield resulting from this increase in liquidity has implications for the exchange rates of recipient countries such as South Africa. As a result, rand appreciation pressures "are expected to persist for some time", and the MPC believes that "The exchange rate therefore remains a downside risk to the inflation outlook."
Recent CPI Undershoot Shaves 0.2pp Off Inflation Trajectory
The final predominant driver of Thursday's decision was the recent lower-than-expected inflation outcomes. As we indicated in previous research (see South Africa: Rate Call Change - We Expect a 50bp Cut, November 1, 2010), at the September MPC meeting, the SARB expected CPI to bottom out at 3.7%Y in 3Q10, before rising to post an average reading of 4.8% in 2011 and an annual rate of 5.1%Y in 4Q12. However, given a July CPI print of 3.7%Y at the time, simple arithmetic would suggest that the SARB's forecast for August and September CPI must have been in the region of 3.6-3.7%Y. Since then, August posted a 3.5%Y out-turn, while September came in at 3.2%Y, taking the quarterly average (and, as a corollary, the 2011 forecast) down by 0.2pp. Against this background, it is clear to us that lower-than-expected outcomes account for 0.2pp of the SARB's 0.5pp downward revision to 2011 CPI (from 4.8% to 4.3%), while a stronger currency and lower administered prices account for the remaining 0.3pp in 2011 and 2012 (from 5.1% to 4.8%).
Easier Money to Boost GDP?
While the MPC highlights a number of downside risks to GDP growth, such as the broadly sideways movement in its leading indicator since April, still-weak private sector gross capital formation, a collapse in manufacturing production, etc, we could not help but notice that the SARB kept its 2010 GDP forecast unchanged at 2.8%, but marginally upgraded its forecasts from 3.2% to 3.3% in 2011 and from 3.5% to 3.6% in 2012. We suspect that this represents the impact of easier money on GDP in the coming two years.
We Believe That Policy Rates Have Bottomed
Looking forward, we believe that this should be the last rate cut in the cycle, for two key reasons. The first is that we believe that CPI has bottomed, and should start grinding higher from here: Our forecasts of 3.3%Y for October and 3.5%Y for November compare with the most recent outcome of 3.2%Y. A turnaround in the inflation trajectory should raise the psychological bar for further policy easing, we think. Further, the SARB's relatively benign CPI outlook for 2012 reduces the scope for further downside revisions going forward, in our view.
It is also clear that the decision to cut rates was driven in part by the recent weakness in manufacturing production. The MPC concedes that industrial action was largely to blame for the sharp fall in output. Nevertheless, it decided not to look through the noise. Looking forward, we expect a technical bounce in the October manufacturing production (i.e., once the impact of the strike falls out of the wash). This should help to lift both the October and November annual readings above the 1.4%Y that was printed in September, thereby obviating the need for a further rate cut.
Risks to Our Call
There are two key risks to our call, however. First is the exchange rate: Our baseline assumption is for USDZAR to close the year at 6.80 before ending 1Q10 marginally weaker at 6.90. Were the currency to appreciate much more than we expect (e.g., USDZAR breaches 6.50), this could have a significant impact on the inflation outlook, thereby allowing the SARB to ease policy further. Second, is the uncertain global environment. Were the problems in the peripheral European countries to spread to the core, thereby impacting negatively on global GDP growth, the SARB could well feel obliged to cut rates again. This is not our base case, however.
Read Full Article »