by Asset Allocation Committee

The Financial Accounts of the U.S. was released last week by the Federal Reserve.  The report is a treasure trove of data about the economy.  This report shows the flow of funds and the balance sheet of the American economy.  Although there is much to look at, here are a few charts that show the underlying health of the economy and the financial system.

First, household debt relative to after-tax income has stabilized.  Essentially, household deleveraging has come to an end but we are not seeing releveraging relative to income.

This is a “good news/bad news” situation.  On the good side, the end of deleveraging will support stronger consumption.  When households are trying to pay back debt, it acts as a dampener on economic growth.  If households begin to add to debt relative to their incomes, growth accelerates.[1]  The bad news is that the deleveraging has ended at historically high levels of debt.  Although this level is probably manageable at current interest rates, spending may prove to be unusually sensitive to interest rates due to the current elevated levels of debt relative to income.

This chart is one we closely monitor; it is net saving by sector.  Net saving in macroeconomics is much like a balance sheet—saving in one sector is always offset by dissaving in another and the sum of the four sectors always equals zero.  What is most disturbing in the data is that household saving has declined and business saving has increased.  In general, business saving funds investment; in a well-functioning economy, businesses should be dissavers.  Rising business saving means less investment and the decline in household saving means that the household sector will become vulnerable to economic weakness.

The last chart of interest is the shares of national income.  We divide national income into what is earned by labor compared to capital.

Since roughly 1990, the labor share has tended to decline and has made new lows in each business cycle.  Capital income has performed in an opposite fashion.  This growing divergence is part of the reason for the rise in populism.  Even with the advanced age of the business cycle and low unemployment, the labor share remains low relative to history.

The Financial Accounts of the U.S. are showing that the expansion is getting a bit old.  Although there is no evidence of a recession on the horizon, the weakness in household saving is a concern.  The end of deleveraging does bode well for the economy in the short run but the continued high level of debt is a worry.  Finally, the current political problems the country is facing will likely require a drop in the share of capital at some point.  That will not be a favorable event for financial assets.

View the PDF

---

[1] Assuming, of course, that all the additional consumption would not be supplied by imports.

by Asset Allocation Committee

In a recent speech,[1] New York FRB President Bill Dudley made the case that the FOMC should continue to reduce monetary stimulus even though inflation remains below target.  His contention is that benign financial conditions in the face of tighter policy are creating distortions in financial markets, resulting in the need for additional policy tightening.

Congress has given the Federal Reserve dual mandates—full employment and low inflation.  The Phillips Curve would suggest that meeting both is often impossible.  The curve postulates that there is a tradeoff between inflation and unemployment, and so meeting one objective probably means missing the other.  Since the 1970s, the Phillips Curve has become increasingly less reliable; globalization and deregulation have led to persistently falling price pressures.  In other words, inflation is falling on its own, and thus monetary policy can mostly focus on full employment.  Based on the current unemployment rate, it is likely that full employment has probably been achieved, although the persistence of weak wage growth would suggest that the Fed should be in no hurry to raise rates.

Although the FOMC has dual mandates, every central bank has the goal of financial stability.  After all, the primary reason for creating a central bank is to build a system for a lender of last resort who will lend to financial institutions during liquidity crises.  The Federal Reserve was created in 1913 in response to the Panic of 1907, which was single-handedly stopped by John Pierpont Morgan (yes, that J.P. Morgan).  President Roosevelt, while relieved that private bankers were able to end the panic, was also worried that relying on this solution in the future was tempting fate.  Thus, he started the debate on creating a U.S. central bank that resulted in the founding of the Federal Reserve six years later.

During financial crises, commercial banks face liquidity problems.  Banks operate by maturity transformation.  They take short-term loans (also known as deposits), repayable on demand, and transform them into less liquid but higher yielding loans.  As long as depositors don’t demand their money en masse, the system works well; cash becomes investable and helps build the economy by providing funds for investment.  However, in a panic, banks may be forced to liquidate good loans to meet the demands from depositors.  This selling can damage the financial system needlessly.  The central bank is designed to lend against these loans so that banks can meet depositor demand and quell the panic.[2]

Essentially, one of the key roles of a central bank is crisis management.  Thus, most central banks have regulatory power to prevent commercial banks from taking excessive risk.  Reserve requirements, capital requirements, bank inspections, stress tests and the general level of interest rates are all used to reduce the likelihood of a panic.  Creating an environment of healthy fear can curb bankers’ “animal instincts” and prevent them from becoming overly optimistic and making aggressive loans.  Unfortunately, if the Federal Reserve is successful in its Congressional mandates, it can prolong the business cycle.  As Hyman Minsky noted,[3] the longer economic conditions appear calm, the greater the likelihood that investors, borrowers and lenders will be inclined to take more risk.  The Minsky Instability Theory postulates that economic actors are more likely to create instability the longer conditions remain stable.

Dudley’s comment about financial stability is worth examining.  On the chart below, we overlay the Chicago FRB National Financial Conditions Index along with the fed funds rate.

The blue line on the chart shows the aforementioned Financial Conditions Index, which measures the level of stress in the financial system.  It is constructed of 105 variables, including the level of interest rates, credit spreads, equity and debt market volatility, delinquencies, borrower and lender surveys, debt and equity issuance, debt levels, equity levels and various commodity prices (including gold).  A rising line indicates increasing financial stress or deteriorating financial conditions.  The red line is the effective fed funds rate.  Until mid-1998, the two series were positively and closely correlated.  When the Fed raised rates, financial stress rose; when the Fed lowered rates, stress declined.  After 1998, the two series became virtually uncorrelated.

We believe there are two factors that changed this relationship.  The first is policy transparency.  Starting in the late 1980s, the Fed became increasingly transparent.  For example, before 1988 the FOMC would meet but issue no statement about what it had decided to do.  Investors and the financial system had to guess whether policy had been changed.  Starting in 1988, the central bank began publishing its target rate.  In the 1990s, it began issuing a statement when rates changed.  Eventually, a statement followed all meetings.  As the FOMC has become more transparent, the correlation between stress and the level of fed funds has changed.  Essentially, the markets now know with a high degree of certainty when rate changes are likely.  This is especially true of tightening.  The FOMC appears to avoid making rate hikes that surprise the market.

The second factor is financial system stability.  From the Great Depression into the 1980s, policymakers put a high premium on system stability at the expense of efficiency.  Bank failures were rare and there were a large number of rather small institutions.  In addition, commercial banks were separated from investment banks.  The drive to improve efficiency in the financial system led to consolidation among commercial banks and a breakdown of the barriers between commercial and investment banks.  Although this made the system more efficient, it also undermined stability.  Thus, when raising rates, the Fed must pay close attention to system stability to prevent crises, which has tended to lead to gradual and measured policies.  This behavior maintains stability…until it doesn’t!

It appears that Dudley would like to return to the pre-1998 period.  We tend to agree with that sentiment.  Monetary policy would be much more effective if financial stress moved directly with changes in policy rates.  However, if our thesis that transparency and industry concentration led to the change in the relationship, it seems highly unlikely that policymakers would reverse those conditions.  Instead, we now live with markets where policymakers have virtually no control over financial conditions; the longer conditions are quiet, the more emboldened investors, lenders and borrowers are likely to become.  And, when financial conditions deteriorate, it seems to require extraordinary measures by central bankers to restore calm.  This means that investors live in a world where financial conditions appear benign most of the time until they are not and then they become quite adverse.  Monetary transparency has probably created distorted financial conditions where risks are hidden and thus encourage risky behavior, suggesting investors should exercise more caution than financial conditions currently signal.

View the PDF

---

[1] https://www.newyorkfed.org/newsevents/speeches/2017/dud170907

[2] The problem for the central bank is determining if a commercial bank faces a solvency or a liquidity crisis.  If the assets of the bank, its loans, are dodgy, the lending against them is probably a mistake and the bank should be liquidated.  On the other hand, if the loans are of good quality, then lending against these loans is a sound way to contain a banking panic.

[3] Minsky, H. (2008). Stabilizing an Unstable Economy (2^nd ed.). New York, NY: McGraw-Hill (originally published 1986).

by Asset Allocation Committee

In our most recent asset allocation rebalancing, we added foreign allocations to our portfolios.  Over the past few years, we had generally avoided allocations to non-U.S. markets in asset allocation portfolios due to two primary concerns.  First, the dollar had been appreciating as a result of an improving U.S. economy and policy divergences between the U.S. and the rest of the world.  The Federal Reserve was raising rates and tapering its balance sheet while the majority of other nations were still adding monetary stimulus.  Second, we have had secular concerns about the stability and attractiveness of foreign investing in a world where the U.S. is seemingly reducing its hegemonic role.

We previously noted that the dollar was deeply undervalued on a purchasing power parity basis and vulnerable to depreciation.  The catalysts for dollar weakness appear to be coming from two sources.  First, the FOMC is moving very slowly to tighten policy while the rest of the world’s central banks are finally withdrawing policy stimulus.  Second, uncertainty surrounding American governance appears to be undermining investor confidence and leading to dollar selling.  In this week’s report, we wanted to update the valuation levels for the dollar against the yen, euro, Canadian dollar and British pound.

This chart shows four purchasing power parity models for the aforementioned currencies.  In all four cases, the dollar was trading “rich” by more than one standard error, and nearly two standard errors from parity in two cases.  Over the past two months, three of the four currencies have begun to appreciate and are indicating some modest improvement in valuation.  However, these models all suggest that the dollar is still overvalued and thus, even with the recent depreciation, the greenback is still overvalued.   Hence, the narrative that a weaker dollar should support further gains in overseas assets remains viable.  If history is any guide, we are still in the early stages of a dollar reversal that should remain in place for the foreseeable future.

At the same time, the secular concerns about the impact of the withdrawal of U.S. hegemony will likely be a bearish factor for overseas investments.  For now, we expect the dollar’s weakness to overshadow concerns over global stability.  But, as some point, possibly in the next couple of years, the dollar will be closer to fair value and the case for foreign investment will be more difficult to justify.

View the PDF

by Asset Allocation Committee

As the FOMC prepares to reduce its balance sheet, it’s a good time to update our views on long-term interest rates.  The chart below shows our current estimate of fair value for the 10-year Treasury.

The model uses fed funds, the 15-year moving average of CPI (an inflation expectations proxy), the yen/dollar exchange rate, oil prices and German bond yields.  The current yield is on fair value.  Assuming the other variables remain steady, the current yield on the 10-year T-note is assuming the FOMC is going to hold rates steady for the foreseeable future.

Is this assumption on the policy rate reasonable?  Currently, fed funds futures don’t reach a 50% chance of a rate hike until the June 2018 meeting, and even by next December the odds of a rate increase are only 67%.  We expect the Yellen FOMC to use balance sheet contraction to placate the hawks on the committee and thus avoid increases in fed funds until it becomes abundantly clear that inflation is rising.

One of the reasons for expanding the Fed’s balance sheet was to lower long-term interest rates.  In reality, the evidence of success is mixed.

This chart shows the 10-year T-note yield with the yearly change in the balance sheet.  The gray bars show official periods of QE.  A zero reading indicates no change in the balance sheet compared to the prior year.  Rates fell in QE1, at least initially, although they did rebound as the recession came to an end in mid-2009.  However, rates generally rose in QE2 and QE3.  Although the Fed was buying longer dated Treasuries, which reduced its supply, it appears the demand may have weakened on fears that QE would trigger inflation.  Thus, a case could be made that reducing the balance sheet would have a similar effect and push rates lower.

Our base case is that reducing the balance sheet will have an asymmetric effect on markets; in other words, it won’t have a significant impact on interest rates, unlike the apparent bearish impact that QE2 and QE3 had on long-term interest rates.  This is because the FOMC is framing the reduction in the balance sheet as “normalization,” whereas QE was designed to be stimulative.  Thus, our analysis suggests that the most important impact of QE was psychological.  However, it is possible that QE did more than just boost sentiment; if so, balance sheet normalization could be bullish for long-duration bonds.

View the PDF

by Asset Allocation Committee

We have previously documented the difference between S&P 500 operating earnings reported by Thomson/Reuters and Standard and Poor’s.  Although both series purport to measure the same thing, there can be rather wide divergences.  These exist due to differences in how unusual events are accounted for; we do often see long periods where the two series are identical but, in this bull market, the Thomson/Reuters data has tended to consistently exceed the S&P numbers.

This chart shows the two series from 1994, with the lower line showing their ratio.  The reason we monitor these divergences is that an elevated ratio has led to bear markets and recessions in two previous instances.  Fortunately, the ratio is narrowing, although the difference in terms of the past four quarters is still notable, with Thomson/Reuters at $125.07 while the S&P is at $116.14.

One reason for the recent change in the ratio could be oil prices.

The drop in oil prices coincided with a widening of the ratio.  Thus, it is possible that S&P treats the impact of falling oil prices on earnings differently than Thomson/Reuters.  If oil prices remain elevated from recent lows, the spread between the two series should also narrow.

This year’s earnings story continues to be the expansion of margins.  Looking at S&P operating earnings compared to GDP, we have seen a solid recovery in margins.  Our margin model has been projecting a recovery and stabilization around the level of 5.5% of GDP.  If that is the case, the growth rate in earnings should slow to the pace of nominal GDP over the next few quarters.

Based on our analysis, we are on pace for a year-end operating earnings number, basis Standard and Poor’s, of $121.50, a 14.3% rise over this series report from last year.  A similar growth number for Thomson/Reuters would put this year’s earnings at $136.10 compared to the current expectation of $131.10.  However, since the gap between the two earnings series is narrowing, current expectations are probably about right as that would be consistent with the current ratio.  If the two narrow further, current expectations may be too high.  Still, in any case, margins remain strong and should offer support for equities.

View the PDF