by Asset Allocation Committee

Equity markets struggled in 2018; although new highs were achieved twice during the year, volatility was elevated compared to 2017.

This is a chart of the CBOE VIX index, a measure of implied volatility from the equity options market.  An elevated VIX means implied volatility is high, which implies a wider dispersion of expected outcomes for future equity index outcomes.  As fear rises, it would make sense for investors to react.  As the chart shows, for most of 2017, the VIX ranged between 10 and 15.  Last year, the range was significantly wider.   As fear rose, investors reacted by accumulating higher levels of cash.

This chart shows the weekly close for the S&P 500 along with the level of retail money market funds.  We have highlighted the 2007-09 recession in gray.  The tan areas show periods when the level of money market funds fell below $920 bn.  In this bull market, we have tended to see the upside momentum wane once money market funds fell below the aforementioned level.  In other words, the market ran out of “dry powder.”

What is occurring now is completely different.  Money market levels are nearly $1.2 trillion.  They are rising in a manner closely resembling what we observed before the Great Recession.  It appears that the level and rise of money market funds is consistent with the financial markets’ expectations of a recession.  If a recession is avoided—if the trade conflict with China cools and the FOMC avoids overtightening—then odds likely favor a recovery in equities.  The retail money market data suggests there is ample liquidity to support a significant rally.  Of course, not all those money market funds will necessarily move back to equities.  Given higher interest rates, some of it could stay in money markets.  But, the key point is that financial markets appear to be in recession mode now.  If recession is avoided in 2019, as we expect, the odds of a significant rally in equities is likely.

View the PDF

by Asset Allocation Committee

Does the Federal Reserve adjust policy for asset prices?  This is perhaps one of the most controversial topics in U.S. monetary policy.  Alan Greenspan faced this issue in the early 1990s.  Both Volcker and Greenspan wanted to focus monetary policy on containing inflation.  But, Larry Lindsey, a Fed governor at the time, noted that if outside forces, such as technology and trade, were keeping inflation down then the Fed could engage in easy monetary policy without the risk of rising price levels.  He warned this could cause asset bubbles.[1]  Greenspan, an adept corporate infighter, prevented Lindsey’s position from gathering any momentum.  But, as the “irrational exuberance” speech showed on December 20, 1996, he became concerned about overheating financial markets.[2]   However, the reaction to the speech led the powerful Greenspan to realize there wasn’t much upside in conducting monetary policy to quell asset bubbles.  Instead, policy evolved to address the aftermath of bubbles.

Still, the idea of low interest rates triggering asset inflation never really went away.  The Great Financial Crisis proved that the costs of cleaning up after a bubble could be considerable.  It was one thing to have a bubble in technology stocks; in general, technology becomes obsolete so quickly that excess capacity in that sector doesn’t have a lasting effect.  On the other hand, a bubble in housing can depress economic activity for years.  Jeremy Stein, a Fed governor from 2012 to 2014, raised concerns about financial excesses.[3]

In the current configuration of the FOMC, shown below, we rate them according to their policy bias (on a 1 to 5 scale, with 1 being the most hawkish and 5 most dovish) and by theoretical inclination.  The latter reflects traditional hawks, characterized by a restrictive view of the Phillips Curve, traditional doves, who have an expansive view of the Phillips Curve, moderates, who make policy based on a variety of factors but tend to be “data-dependent” (in practice, atheoritical and not tied to the Phillips Curve) and financial asset-sensitive.  The table below shows the breakdown.  The number shows policy bias based on our analysis of comments and voting patterns.  The colors show what we view as their theoretical background.  Among the voters this year, the average is nearly 3, suggesting a moderate voting bloc.  This year, there is only one dove and one hawk, five moderates and three financial market-sensitive voters.  The doves tend to raise rates reluctantly; hawks tend to cut rates with the same distaste.  Moderates are mostly a diverse group from a theoretical perspective.  For our purposes, the important difference of this group compared to the traditional hawks and doves is skepticism about the Phillips Curve.  These voters tend to watch trends in the overall economy and make policy decisions.  Interestingly enough, three of the governors appointed by President Trump have been moderates and he also promoted Jerome Powell to chair of the FOMC.  For a president who seems to prefer doves, he has been steered into appointing moderates.  Finally, there are three members who, in the comments, seem much attuned to the behavior of financial markets.  Governor Brainard has voted as a dove but has expressed concern about market overheating and has used that position to support recent rate hikes.

This year, we have three voters we dub as “financial-sensitive.”  Thus, financial market behavior may be important to the path of policy this year.

However, as Greenspan noted, it’s hard in real time to determine whether an asset market is in a bubble.  And, it can be equally difficult to determine whether the cost of raising rates to prevent the bubble is less expensive than addressing the aftermath.  The key problem with asset bubbles is that they leads to malinvestment.  In a long-lasting asset, that can mean years of technical inefficiency because capacity can’t be fully utilized.  Thus, a housing bubble can lead to too much real estate that can take years to absorb; cutting interest rates can help slow the inevitable decline in prices but may actually expend the period necessary to balance the market.  On the other hand, a bubble in wheat lasts one growing season and policymakers shouldn’t bother to address the problem.

In addition, it would be politically explosive for policymakers to raise rates solely because equity or home prices have risen “excessively.”  The backlash would threaten central bank independence.  Thus, if the Fed is worried about an asset bubble, it would need some measure other than valuation to raise rates.

One possibility we have examined recently is volatility.  Does the FOMC adjust rates based on the equity market VIX?  There appears to be some evidence that policymakers may be sensitive to market volatility.

This chart shows the weekly fed funds target with the 12-week average of the CBOE VIX index.  We have placed a bold line at 20 for the VIX.  Since the late 1990s, we note that the FOMC was inclined to keep lowering rates with a VIX above 20; a reading under 20 would tend to support policy tightening.  So, in 2002, Chair Greenspan kept cutting rates even though the economy was in clear recovery.  It may have been due to perceptions that investor sentiment was overly negative.  The 2004-06 tightening cycle occurred with a VIX persistently below 20.  In fact, rate cuts seemed to occur as the VIX rose.  We also note that the 2016 pause occurred after the VIX rose back above 20, and tapering was announced in 2016 after a prolonged period of a low VIX.  The current pause coincides with the recent lift in volatility.

We also examined adding the VIX to the Mankiw Rule model variations.  What we found is that the index is statistically significant in three of the variations and the correct sign in two.  However, in the variations it did correctly affect, it didn’t necessarily improve the forecasting accuracy by more than 10 bps.  This performance suggests that the VIX may have an impact on policy but the Phillips Curve variables, labor market data and inflation, are still more important.  However, the hard part to divine is the impact of the VIX on the moderate voters.  Even if all of the market-sensitive members pay attention to the VIX, the moderates may only pay attention at extremes.

Therefore, in conclusion, we can probably say the following—when the VIX is below 20, the Fed is probably more likely to consider tightening policy.  A reading above 20 may lead to a pause or could encourage further easing.  However, the relationship isn’t precise, which suggests the traditional hawks and doves don’t pay much attention to market volatility.  The VIX may be the way that market-sensitive FOMC members can incorporate financial markets into their policy decisions without overtly targeting valuations or returns.

View the PDF

---

[1] Mallaby, Sebastian. (2016). The Man Who Knew: The Life and Times of Alan Greenspan. New York, NY: Penguin Books. (pp. 435-36.)

[2] Ibid, pp. 504-506.

[3] https://fraser.stlouisfed.org/title/1163/item/2372 and https://fraser.stlouisfed.org/title/1163/item/476707

by Asset Allocation Committee

Quantitative easing (QE) was an element of unconventional monetary policy that emerged from the Great Financial Crisis.  When the Federal Reserve lowered the fed funds target to zero (known as the “zero interest rate policy,” or ZIRP), policymakers decided that taking the policy rate below zero would not further stimulate the economy.  Once interest rates fall below zero, idle cash generates a return and there was fear that negative interest rates would cause cash hoarding, which was exactly what the Fed wanted to avoid.  However, policymakers also feared that signaling they lacked additional tools to support the economy could trigger a panic in financial markets.  Thus, they created an additional tool, QE, which bought Treasuries and mortgages from the financial system and pushed additional cash into the economy.

The policy was controversial.  Some feared it would trigger inflation as the high levels of cash would “inevitably” cause rising prices.  Others argued that the process would distort financial markets.  But, in the end, the effects of QE were difficult to parse.  For example, one of the Fed’s arguments for QE was that it would lower long-term interest rates.  In fact, at times, it seemed to have precisely the opposite effect.

This chart shows the 10-year T-note yield.  The gray bars show periods of QE.  In all three events, yields rose at the onset of the balance sheet expansion.  Yields eventually fell during QE2, although that decline was probably more due to turmoil in Europe.  Why did yields rise when the Fed was reducing the supply of bonds?  Most likely, investors worried that QE would be successful in bringing about inflation and thus demand was adversely affected.  Anticipation of the end of QE always led to a drop in yields.

What about all that money that was injected into the system?  Much of it remained on bank balance sheets in the form of excess reserves.

Using the equation of exchange (money supply x velocity = price x quantity), the increase in the money supply mostly led to a sharp drop in velocity, instead of boosting prices or quantity.

So, if QE mostly sat idle on bank balance sheets then the withdrawal of QE should not have had much of a real impact on the economy.  And, that is likely true.  However, this isn’t to say that QE had no effect.  As we have noted in the past, there has been a close correlation between the S&P and the Fed’s balance sheet.

The chart on the left shows the original balance sheet model; note that after Trump’s election in 2016 the S&P rose much more than the model would have projected.  To account for this move, we built a variable that adjusted for expectations surrounding corporate tax cuts.  The additional variable improved the model dramatically; however, assuming no further reductions in corporate tax rates, the market is once again tracking the balance sheet.

Essentially, it appears that the impact of QE was mostly psychological; it signaled to investors that the Fed was supportive.  Interestingly enough, the withdrawal appears to be adversely affecting investor sentiment.  It is hard to determine how much, simply because there are other factors, such as trade conflicts and FOMC criticism, which are also affecting sentiment.  However, this data suggests that a pause in the balance sheet reduction would likely improve investor sentiment.

View the PDF

by Asset Allocation Committee

(N.B.  This is the last Asset Allocation Weekly for 2018.  Have a Merry Christmas and Happy New Year.  The next report will be published January 4, 2019.)

The U.S. economy is performing in line with the rest of the world.

This chart shows the yearly change in U.S. and world ex-U.S. GDP.  The lower line on the chart shows the difference.  On average, U.S. growth is usually 0.61% lower than world growth and the world exceeds U.S. growth 70% of the time (with data since 1981).  This situation isn’t a huge surprise; the U.S. is the world’s largest economy and smaller economies can grow faster more easily.  The lower line shows that U.S. growth has been gaining on world growth for the past two years.

The net number does coincide with dollar cycles.

When the growth differential is below average (implying stronger world growth relative to the U.S.), the JPM dollar index averages 107.21.  When growth exceeds average, the index averages 113.14.  We are projecting slower growth in the U.S. next year, around 2.7%, which is essentially average growth.  If world growth also holds near average, around 3.3%, it would be reasonable to expect the dollar to weaken from current levels.  The U.S. growth surge in 2015 led to a strengthening dollar and this year’s rally was partly due to the lift in U.S. growth due to fiscal stimulus.  As that wanes, relative growth should favor the world, which will support a softer greenback.  In general, a weaker dollar will tend to support commodities and foreign equities, especially emerging markets.   We expect those assets to perform better in 2019.

View the PDF