V- or M-shaped correction?

Last week I mentioned that there are few “V-shaped” corrections and plenty with a “W-shape”. There are also a few with an “M-shape”, leading to a major market sell-off. Here are some examples on Dow Jones Industrial Average.

2001 is the only good example I can find of a V-shaped correction.

Dow Jones Industrial Average

It rolled over later in 2002 into a more conventional W-shape bottom with several tests of support at 7500.

Dow Jones Industrial Average

This was followed by the banking crisis of 2008 which started with an M-shape in 2007. Successive false breaks above resistance (orange arrows) were followed by breach of support (red arrows)…before Lehman Bros filing for bankruptcy on September 15 led to a major capitulation.

Dow Jones Industrial Average

2011 is nowadays considered a secondary movement but at the time caused widespread alarm. Starting with an M-shaped top, it broke support in August before forming a W-shaped bottom with several tests of support at 11000.

Dow Jones Industrial Average

2015 was a more conventional W-shape precipitated by falling oil prices.

Dow Jones Industrial Average

Now, in 2018, we have the makings of either a W-shaped correction or an M-shaped reversal. The false break above resistance at 26500 is definitely bearish but was followed by a bullish higher low at 24000.

Dow Jones Industrial Average

There are three possible options:

  1. Completion of a W-shape correction, with breakout above 27000;
  2. An M-shaped reversal, with a fall below 23500; or
  3. A lengthy consolidation reflecting uncertainty, as in 1999 to 2001.

Dow Jones Industrial Average

At this stage, option 1 is most likely. Buybacks and strong Q3 earnings are likely to counter bearish sentiment.

That would change if we see:

A negative yield curve, where the 3-month T-bill rate crosses above 10-year Treasury yields;

Yield Differential

Rising troughs above 1% on the S&P 500 21-day Volatility Index; or

S&P 500

Bellwether transport stock Fedex follows-through below support at 210.

Fedex

Remember that there is nothing stable in human affairs; therefore avoid undue elation in prosperity, or undue depression in adversity.

~ Socrates

Treasury yields confirm bond bear market

10-Year Treasury yields respected their new support level at 3.00%, confirming a primary advance.

10-year Treasury Yield

Breakout above 3.00% also completes a double-bottom reversal, signaling the end of a three-decade-long secular bull market in bonds.

LT 10-year Treasury Yield

The yield differential between 10-year and 3-month Treasuries is declining but a flat yield curve does not warn of a recession. Only if the yield differential crosses below zero, with short-term yields rising faster than long-term, will there be a recession warning.

Real returns on long-term bonds — the gap between the green and blue lines below — remain near record lows.

1981 to 2018: 10-Year Treasury Yields and GDP Implicit Price Deflator

Only if the gap widens (real returns rise significantly) are we likely to see downward pressure on stock valuations, with falling price-earnings multiples.

How will a bond bear market affect stocks?

10-Year Treasury yields broke out of their triangular consolidation at 3.00%, while the Trend Index recovered above zero signaling a fresh advance.

10-year Treasury Yield

Importance of resistance at 3.00% is best illustrated on a long-term monthly chart. Yields declined for more than three decades (since 1981) in a bond bull market but the rise above 3.00% completes a double-bottom reversal, warning of rising yields and a bond bear market. Target for the advance is 4.50%.

10-year Treasury Yield

The yield differential between 10-year and 3-month Treasuries has declined since 2010, prompting discussion as to whether a flat yield curve will cause a recession.  Interesting that the yield differential recovered almost 20 basis points in September, with long-term yields rising faster than short-term. Penetration of the descending trendline would suggest that an imminent negative yield curve is unlikely.

10-year Treasury Yield

How would a bond bear market affect stocks?

Capital losses from rising yields on long-maturity bonds would increase demand for shorter maturities, driving down short-term yields and causing a steeper yield curve. A bullish sign for stocks.

Inflation is low and the rise in long-term yields is likely to be gradual. Another bullish sign.

The last bond bear market lasted from the early 1950s to a peak in September 1981. Higher interest rates were driven by rising inflation ( indicated below by percentage change in the GDP implicit price deflator). The 1975 spike in inflation was caused by the OPEC oil embargo in retaliation for US support of Israel during the 1973 Yom Kippur war.

1950 to 1981: 10-Year Treasury Yields and GDP Implicit Price Deflator

Stock prices continued to climb during the bond bear market, apart from a 1973 – 1974 setback, but the Price-Earnings ratio fell sharply in ’73-’74 and only recovered 10 years later, in the mid-1980s.

1950 to 1981: S&P 500 and PE Ratio

Alarmists may jump to the conclusion that a bond bear market would lead to a similar massive fall in earnings multiples but there were other factors in play in 1975 to 1985.

First, crude prices spiked after the OPEC oil embargo and only retreated in the mid-1980s.

1960 to 1985: West Texas Intermediate Crude prices

The rise of Japan also threatened US dominance in global markets.

1960 to 1985: Nikkei 225 Index

We should rather examine the period prior to 1973 as indicative of a typical bond bear market. The S&P 500 Price-Earnings ratio was largely unaffected by rising yields. Real interest rates actually decreased during the period, with the gap between 10-year yields and the inflation rate only widening near the 1981 peak.

At present, real interest rates are near record lows.

1981 to 2018: 10-Year Treasury Yields and GDP Implicit Price Deflator

We can expect real interest rates to rise over time but that is unlikely to have a significant impact on earnings multiples — unless there is a strong surge in long-term yields ahead of inflation.

 

Does the yield curve warn of a recession?

There has been talk in recent months about the narrowing yield curve and how this warns of a coming recession, normally accompanied by a graph of the 10-year/2-year Treasury spread which fell to 0.22% at the end of August 2018.

Yield Differential 10Year minus 2Year

I have always used the 10-year minus the 3-month Treasury spread to indicate the slope of the yield curve but, although this shows a higher spread of 0.71%, both warn that the yield curve is flattening.

Yield Differential 10Year minus 3Month

Is this cause for alarm?

First of all, what is the yield curve? It is the plot of yields on Treasuries against their maturities. Long maturity bonds normally have higher yields than short-term bills, to compensate for the increased risk (primarily of interest rate changes). If you tie your money up for longer, you expect a higher return. That is a rising yield curve.

A steep yield curve is a major source of profit to banks as their funding is mostly short-term while they charge long-term rates to borrowers, pocketing the interest spread.

The Fed sometimes intervenes in the market, however, restricting the flow of money to the economy, to curb inflation. Short-term rates then rise faster than long-term rates and the yield curve may invert — referred to as a negative yield curve.

At present we are witnessing a flattening yield curve, as short-term rates rise close to long-term rates.

A recent paper from Michael D. Bauer and Thomas M. Mertens at the San Francisco Fed concludes that a narrow yield differential has zero predictive ability of future recessions:

In light of the evidence on its predictive power for recessions, the recent evolution of the yield curve suggests that recession risk might be rising. Still, the flattening yield curve provides no sign of an impending recession. First, the evidence suggests that recession predictions based on the yield curve require an inversion (Bauer and Mertens 2018); no matter which term spread is used to measure its shape, the yield curve is not yet inverted. Second, the most reliable summary measure of the shape of the yield curve, the ten-year–three-month spread, is nearly 1 percentage point away from an inversion.

I was pleased to see that Bauer and Mehrtens find the 10-year/3-month Treasury spread more reliable than other spreads in predicting a recession within 12 months, with 89% predictive accuracy. They also refer to another study that came to a similar conclusion:

Engstrom and Sharpe found that their short-term spread statistically dominated the 10y–2y spread, and our findings are consistent with this result.

But both studies conclude that a negative yield curve (when the yield differential is below zero) is a reliable predictor of recessions. And Bauer and Mehrtens observe that, while the 10 year/2 year spread is less accurate, it is still a reliable predictor.

Are we just 22 basis points away from a recession warning? Let’s weigh up the evidence.

First, a negative yield curve is a reliable predictor of recessions. In the last 60 years, every time the 10-year/3-month spread has crossed below zero, a recession has followed within 12 months. There is one arguable exception. In 1966 the yield differential crossed below zero, the S&P 500 fell 22% and the NBER declared a recession, but they (the NBER) later changed their mind and airbrushed it out of history.

Yield Differential 10Year minus 3Month

Second, while there is strong correlation between the yield curve and recessions, the exact relationship is unclear.

The most convincing explanation is that bank interest margins are squeezed when the yield curve inverts. When it is no longer profitable for banks to borrow short and lend long, they restrict the flow of new credit. Credit is the lifeblood of the economy and activity slows.

That was clearly the case in the lead up to the 2008 crash, but why are net interest margins of major US banks now widening?

Bank Net Interest Margins

The flow of credit also slowed markedly before the 1990/1991 recession but did not ahead of the last two recessions.

Bank Net Interest Margins

And growth in the broad money supply — zero maturity money (MZM) plus time deposits — accelerated ahead of the Dotcom crash and 2008 banking crisis.

Broad Money Supply: MZM plus Time Deposits

Third, consider the Wicksell spread. Swedish economist Knut Wicksell argued in his 1898 work Interest and Prices that the economy expands when return on capital is higher than the cost of capital, with new investment funded by credit, and it contracts when the expected return on capital is below the cost of capital.

I was first introduced to Wicksell by Neils Jensen, who uses the Baa corporate bond yield as a proxy for the cost of capital and nominal GDP growth for the return on capital. Neils argues that the economy is near equilibrium when the Wicksell spread is about 2.0% — when return on capital is 2.0% higher than the cost of capital.

Wicksell Spread: Nominal GDP Growth compared to Baa Corporate Bond Yield

The above graph shows that 1960 to 1980 was clearly expansionary, with nominal GDP growth exceeding the cost of capital (Baa corporate bond yield). But the last almost four decades were the opposite, with the cost of capital mostly higher than the return on capital. Only recently has this reversed, suggesting a new expansionary phase.

One could argue that low-grade investment bond yields are a poor proxy for the cost of capital, with rising access to equity markets in recent decades. Also that nominal GDP growth rate is a poor proxy for return on capital. If we take the S&P 500, the traditional method of calculating cost of equity is the current dividend yield (1.8%) plus the dividend growth rate (8.0%), giving a 9.8% cost of capital. If we take the current S&P 500 earnings yield of 4.0% (the inverse of the P/E ratio) plus the earnings growth rate of 15.1% as the return on capital (19.1%), it far exceeds the cost of capital. You can understand why growth is soaring.

New capital formation is starting to recover.

New Capital Formation

Fourth, Fed actions over the last decade have distorted the yield curve. More than $3.5 trillion of Treasuries and mortgage-backed securities (MBS) were purchased as part of the Fed’s quantitative easing (QE) strategy, to drive down long-term interest rates. In 2011 to 2012, the Fed also implemented Operation Twist — buying longer-term Treasuries while simultaneously selling shorter-dated issues it already held — to further bring down long-term interest rates. Long-term rates are still affected by this.

In addition, Fed efforts to shrink their balance sheet may further distort the yield curve. The Fed has indicated that it will not sell Treasuries that it holds but will not reinvest the full amount received from investments that mature. If we consider that short-term Treasuries are far more likely to mature, the result could be that the maturity profile of the Fed’s Treasury portfolio is getting longer — a further extension of Operation Twist by stealth.

Conclusion

A flat yield curve does not warn of a coming recession. A negative yield curve does. Both the 10-year/2-year and 10-year/3-month Treasury spreads are reliable predictors of a recession within 12 months, but the 10-year/3-month spread is more accurate.

The correlation between the yield curve and recessions is strong but the actual relationship between the two is more obscure. Links between the yield curve, bank net interest margins, bank credit growth and broad money supply growth are more tenuous, with lower correlation.

Also, return on capital is rising while cost of capital remains low, fueling strong capital formation. The economy is starting to grow.

Fed actions, through QE, Operation Twist, and even possibly steps to unwind its balance sheet, have suppressed long-term interest rates and distorted the yield curve. While the yield curve is still an important indicator, we should be careful of taking its signals at face value without corroborating evidence.

Lastly, we also need to consider the psychological impact. If the market believes that a negative yield curve is followed by a recession, it most likely will be. Beliefs lead to actions, and actions influence outcomes.

Treat yield curve signals with a great deal of respect, and be very wary of how the market reacts, but don’t mindlessly follow its signals without corroboration. The economy may well be entering a new growth spurt, with all its inherent dangers — and rewards.

I contend that financial markets never reflect the underlying reality accurately; they always distort it in some way or another and the distortions find expression in market prices. Those distortions can, occasionally, find ways to affect the fundamentals that market prices are supposed to reflect.

~ George Soros

Low inflation risk keeps yield curve safe

The Fed is advancing interest rates at a measured pace, with the objective of restoring balance in financial markets rather than to curbing inflationary pressures. Only if inflation spikes is the Fed likely to adopt a restrictive stance.

Elliot Clarke from Westpac sums up the FOMC (Fed Open Market Committee) view from their latest minutes:

Beginning with inflation, whereas the market has recently been concerned that inflation may be getting away from the FOMC (given annual CPI inflation at 2.5%yr and persistent strength in the oil price), the Committee is unperturbed.

Instead of the CPI, the FOMC’s benchmark remains PCE inflation, which is currently 2.0%yr on a headline basis and 1.9%yr for core…..

To see upside inflation risks build, a stronger wage inflation pulse is necessary. At present the employment cost index is only reporting “a gradual pickup in wage increases”, and the signal from other wage measures is “less clear”. Two other important considerations for the pass through of wages to activity and thus inflation is that real hourly earnings growth is currently flat and the savings rate near historic lows. The capacity of households to boost consumption and thus inflation is therefore very limited.

Hourly wage rates are growing at a gradual pace.

Hourly Wage Rate Growth

Personal savings are low.

Personal Savings

And credit growth is modest.

Credit Growth

So not much sign of inflationary pressure.

….Turning to financial conditions, as yet there is no concern of them becoming an impediment to growth or policy. The 10yr yield has moved back to the highs of 2013, but the US dollar has only partly retraced its 2017 depreciation. Further, asset markets remain near recent highs.

Equally significant however is the reference to being nearer neutral and a clear desire to keep the yield curve’s positive slope…..

We do not believe that the yield curve will invert in this instance, in part because higher deficits should see the term premium rise. However, the curve will remain comparatively flat versus history, restricting both the timing and the scale of further rate hikes. This is a key justification for both the market’s and our own view of only two further hikes in 2018 and two more in 2019 – a stark contrast to the FOMC’s seven hikes to end-2020.

Yield Differential

A negative yield curve — when 10-year minus 3-month Treasury yields falls below zero — would give a strong recession warning. But the yield curve is only likely to invert if the Fed steps up interest rate increases. With little sign of rising inflationary pressure at present, the prospect seems remote.

Life left in US stocks

According to market pundits, the latest stock sell-off was fueled by concerns over rising bond yields and slowing growth for Caterpillar (CAT).

From CNBC:

….Caterpillar shares reversed lower during the call, when Chief Financial Officer Brad Halverson said first-quarter adjusted profits per share will be the highest for the year because of increased investment later in 2018.

“We expect the targeted investments for future growth to be higher over the remaining three quarters,” Halverson said. “The outlook assumes that first-quarter adjusted profit per share will be the high-water mark for the year.”

Caterpillar (CAT)

The stock fell 6.2% on Wednesday, ignoring the earnings report:

In the earnings report, the Illinois-based machinery manufacturer raised its 2018 profit outlook by $2 a share over the previous quarter, to a range of $10.25 to $11.25 per share. The rosier guidance exceeds a Reuters analyst survey that expected a range of $8.39 to $10.60 a share. The company cited better-than-expected sales volume as the main driver of its improved full-year guidance.

Since when has “better-than-expected sales volume,” upward earnings revision and increased new investment been a bear signal? The market is unusually jittery at present, focusing on any semblance of bad news and ignoring the good.

Even concern over rising bond yields is nothing new.

10-Year Treasury Yields

10-Year Treasury yields are testing resistance at 3.0%. Breakout would complete a double-bottom reversal, warning of a bear market in bonds as yields rise. But rising long-term rates are not bad news for stocks, especially when off a low base as at present. I would go so far as to say that, over the last 20 years, rising 10-year yields have been bullish for stocks. The chart below compares annual percentage change in 10-year Treasury yields and the Russell 3000 Total Market index.

10-Year Treasury Yields and Russell 3000 Index 12-Month Rate of Change

There is plenty more good news that the market seems to be ignoring.

First quarter 2018 corporate earnings have so far impressed. According to S&P Indices, 117 stocks in the S&P 500 had reported results by the morning of April 24th. Of those, 91 (77.8%) beat, 10 (8.5%) met and 16 (13.7%) missed their estimates. Misses are largely concentrated in Materials ( 3 of 5), Industrials (4 of 26) and Consumer Discretionary sectors (5 of 13).

Freight activity remains strong, signaling a reviving economy.

S&P 500

Wages growth remains tame, with average hourly earnings of production and non-supervisory employees increasing at an annual rate of 2.42%. Growth above 3.0% would warn that underlying inflation is rising and the Fed will be forced to tighten monetary policy. But that does not appear imminent.

S&P 500

Muted wages growth allowed corporate profits (the blue line below) to rebound after a threatened down-turn.

S&P 500

Consumption has recovered. Per capita consumption of non-durable goods is recovering after a flat spot in 2017, consumption of durable goods has been rising since 2016, while services remain strong.

S&P 500

In financial markets, risk premiums on corporate bonds (Baa minus Treasuries) have declined to below 2.0%, suggesting a healthy credit outlook.

S&P 500

Bank credit is recovering after faltering in 2017.

S&P 500

The yield curve is flattening as the Fed gradually raises interest rates. A flat yield curve is not a threat. Only if it inverts, when the yield differential (gray line on the chart below) falls below zero, is the economy at risk of falling into a recession. Growth in the money stock (green MZM line on the chart below) has slowed but remains healthy.

S&P 500

The Fed has committed to shrinking its $4 trillion investment in Treasuries and mortgage-backed securities (MBS) run up by quantitative easing (QE) between 2009 and 2014. So far the decline has had no impact on financial markets as bank excess reserves on deposit at the Fed are declining at a similar rate. The effect is that net assets (Fed Assets minus Excess Reserves) are holding steady at $2.4 trillion.

S&P 500

The Philadelphia Fed’s Leading Index remains healthy at above 1.0 percent.

S&P 500

And our estimate of real GDP is rising (2.14% in March 2018), suggesting that the economy is recovering from its flat spot in 2016/2017.

S&P 500

Valuations are high and investors are jittery but the bull market still appears to have further to run.

Leading Index gives early warning

One of the better composite indicators in the US, the Leading Index from the Philadelphia Fed, points to a slow-down in the US economy. A dip below 1.0% is often early, as in July 2000 and May 2006, but serves as a reliable warning of an economic slow-down.

Leading Index for the United States

The Leading Index predicts the six-month growth rate of the Philadelphia Fed Coincident Index. In addition to the Coincident Index, it includes variables that lead the economy: housing permits (1 to 4 units), initial unemployment insurance claims, delivery times from the ISM manufacturing survey, and the interest rate spread between the 10-year Treasury bond and the 3-month Treasury bill.

The Coincident Index combines four indicators: nonfarm payroll employment, the unemployment rate, average hours worked in manufacturing and wages and salaries.

Coincident Index for the United States

The Leading Index signal does seem early. Low corporate bond spreads and VIX near record lows continue to indicate low market risk, typical of a bull market.

Corporate Bond Spreads and VIX

Monetary policy remains accomodative, with money stock growing at close to 5% p.a. (MZM = cash in circulation, travelers checks, money market funds and deposits with zero maturity).

MZM and Yield Differential

The yield curve has flattened, with the spread between 10-year and 3-month Treasuries falling to 1.0% on the above graph. That is what one would expect when the Fed hikes interest rates in a low inflation environment: short-term rates will rise faster than long-term rates. But a negative yield curve, where short-term rates are higher than long-term rates, is a reliable predictor of recessions in the US economy. Each time the yield differential on the above graph crossed below zero in the last 50 years, a recession has followed within 12 months.

Underlying inflation remains low, with average hourly earnings growth below 2.5% p.a., and the Fed should be careful about single-mindedly raising interest rates without considering the yield curve.

Annual Growth in Average Hourly Earnings

The bull market continues but investors need to keep a weather eye on interest rates and the yield curve.

VIX hits record low

The CBOE Volatility Index (VIX) made a new low of 9.30 indicating record low levels of stock volatility. High levels of stock buybacks and large ETF fund inflows may both have contributed, but this is only the third time in its 27-year history that index has broken below 10%. The first was in late 1993. The second, in late 2006, was followed a year later by a massive market snap-back. This time is no different. Volatility is unlikely to remain at such low levels and eventually we will see a market down-turn, accompanied by high volatility, but there is no crystal ball that can tell us whether this will be in one year or five.

CBOE Volatility Index (VIX)

Corporate bond spreads are also falling, with the spread between lowest investment grade Baa (10-year) and equivalent Treasury yields at their lowest point since 2008.

Corporate Bond Spreads

Source: St Louis Fed & Moody’s

The yield curve is flattening but remains comfortably above a flat or negative yield curve when
the yield differential (10-year minus 3-month yields) falls below zero. A negative yield curve is a reliable warning of recession within 12 months.

Yield Differential

Source: St Louis Fed

The Freight Transportation Services Index displays a steady increase in economic activity.

Freight Transportation Services Index

Source: St Louis Fed & US Bureau of the Census

And the S&P 500 continues its advance towards 2500.

S&P 500

Target 2400 + ( 2400 – 2300 )

Bond spreads bullish for US, less so Australia

Yield Curve

The yield curve is one of the best predictors of US economic recessions. Every time the yield curve has turned negative in the last fifty years, a recession has followed.

First of all, what is a yield curve? It is the plot of yields on bonds, normally Treasuries, against their maturities. Long maturity bonds are expected to have higher yields than short-term bills, to compensate for the increased risk (primarily of interest rate changes). If you tie your money up for longer, you would expect a higher return. Hence a rising yield curve.

A rising yield curve is a major source of profit to the banks as their funding is mostly short-term while they charge long-term rates to borrowers, pocketing a healthy interest margin.

When the Fed steps into the market, however, restricting the flow of money into the economy, then short-term rates rise faster than long-term rates and the yield curve can invert (referred to as a negative yield curve).

Bank interest margins are squeezed — it is no longer profitable to borrow short and lend long — and they restrict the flow of new credit.

Credit is the lifeblood of the economy and activity slows.

The chart below compares US recessions to the yield differential: the difference between 10-year Treasury yields and the yield on 3-month T-bills. The yield differential falls below zero when 3-month T-bills yield more than 10-year T-notes.

Yield Differential: 10-year Treasury yields minus 3-month T-bills

You can see that every time the yield differential dips below zero it is followed by a gray bar indicating a recession. There is one exception: the phantom recession of 1966 when the S&P 500 fell 22%. This was originally certified as a recession by the NBER but they later changed their mind and airbrushed it out of history.

You can also see that the yield differential is declining at present but, at 2.0%, it is a long way from a flat or negative yield curve. This supports my argument last week that current Fed rate hikes are more about normalizing interest rates than about monetary tightening.

That could change in the future but at present the bull market still appears to have plenty in the tank.

Corporate Bond Spreads

Corporate bond spreads — the yield difference between high-grade corporate bonds and the risk-free Treasury rate — are another useful indicator of the state of the economy.

Wide bond spreads indicate increased risk of corporate default. Investors are concerned about the state of the economy and demand a higher premium for taking credit risk.

Narrow spreads suggest that credit premiums are low and confidence in the economy is good.

If we examine the chart below, bond spreads are declining, indicating confidence in the US economy, with even the lowest investment grade BBB dipping below 150 basis points (or 1.50%). This is synonymous with a bull market.

US Bond Spreads

Australian corporate bond spreads are higher than the US, with BBB still at 200 bps. They have also declined over the last year but seem to be trending upward from their 2013 low. This is not conclusive as the current trough is not yet complete, but a higher low would warn that credit risk is rising.

Australian Bond Spreads

Only when the tide goes out do you discover who’s been swimming naked.

~ Warren Buffett

Flattening yield curve & low bank interest margins

The Yield Differential, calculated by subtracting 3-month from 10-year Treasury Yields, is trending lower. This warns that the yield curve is flattening but we are still above the danger area below 1.0 percent.

Yield Differential: 10-Year minus 3-Month Yields

A flat yield curve squeezes bank interest margins and often precedes a credit contraction.

Large US Banks: Net Interest Margins

But there is little sign of slowing credit growth so far.

US Bank Loans & Leases: Annual Growth

The St Louis Fed Financial Stress Index (STLFSI) continues to indicate low market stress.

St Louis Fed Financial Stress Index

The STLFSI measures the degree of financial stress in the markets and is constructed from 18 weekly data series: seven interest rate series, six yield spreads and five other indicators. Each of these variables captures some aspect of financial stress. Accordingly, as the level of financial stress in the economy changes, the data series are likely to move together.