Summary

Team view
The asset classes or sectors mentioned herein are a reflection of the portfolio manager’s current view. These comments should not be considered as investment research or recommendation advice. Any prediction, projection or forecast on sectors, the economy and/or the market trends is not necessarily indicative of their future state or likely performances.

One consistent narrative that we have heard over the past few years is that interest rates are too low and will struggle to go lower. Despite this, every time the equity market has found itself in trouble, central banks have eased policy allowing bonds to play the diversifying role that they always have. Given the Reserve Bank of Australia has cut the cash rate to 0.25%, it raises a more pointed question: “Can bonds offer protection again?”

Australian rates

The simplest answer to this questions is: it depends. It depends on the options available to the Reserve Bank of Australia (RBA) and whether they will exercise them, as they have shown reluctance to want to move down this path. Some of the policy measures may seem extreme, while others mild, but it is the use of these policies that will allow yields to move lower. To get to the heart of this matter, I will examine our preferred relationships to show how yields could move and focus in on an equally important question: would the RBA do it?

Chart 1 shows the yield differential between the cash rate and the Australian Composite Bond Index. You can see that the cash rate is the key driver of yields, making RBA policy extremely important. The Composite index very rarely trades under cash, so if the RBA believes the cash rate has hit its lower bound, it’s understandable that people will question duration.

Chart 1 Cash rate and bond yields

Before we dive too deeply into the analysis, there are a few key observations to call out about rates and spreads:

  1. Typically, the Composite Yield hovers between -0.5% and +1.0% around cash. This provides a good range for estimating yields.
  2. The spread between cash and the Composite typically falls to zero just before the first RBA cut in a cycle. This means if we know the RBA is going to cut, we should be overweight before it happens.
  3. The spread typically struggles through 100bps over cash. So, if the RBA is not going to be moving cash, we have a good estimate of where yields should top out if the market sells off.
  4. The spread usually steepens up just before hikes. So, if we believe the next move in the cash rate could be up, then that is our signal to become concerned about duration.

With this information in mind, we can focus on the most important questions which will help determine whether bonds will provide protection against an equity market sell-off:

  • Do we think the RBA could take rates negative (or is that completely off the table)?
  • Can the RBA expand their current quantitative easing (QE) program or introduce other QE tools?

Given that we are concerned about protection, rather than a mid-point forecast, we need to consider these questions in the context of whether the RBA would be more inclined to use these policies if the equity market fell 20%, as the past 10 years have shown that central bank reactions can be larger when equities are falling.

While the analysis I present isn’t intended to represent what I think is the most likely scenario, it does show how the bond market could react if things go wrong.

Will the RBA ease again?

The starting point here is to look at what the RBA is saying and question whether they could ease again, either through QE or by cutting rates. At Nikko AM, we have been discussing this for some time, but my feeling is that the RBA is starting to show signs of dovishness and is thinking about the ways it can ease conditions.

It goes without saying that the economic information is as bad as we have ever seen it. From a historical perspective, this sort of economic shock would have seen rates fall four to five per cent, but as the cash rate started the crisis at only one per cent, we haven’t done anywhere near that yet.

Here is our Australian Economic Bias Indicator, which shows economic conditions at their lowest point in the 25 years that we have run the index. The red dots indicated cash rate moves and 2008 shows an example of how the cash rate fell in similar conditions—multiple 100-point cuts for good measure.

Chart 2 Australian monetary bias long-term indicator

If the RBA had not hit what they thought was the lower bound, we believe they would already have cut rates further. Hence, the economic conditions and lack of conventional policies mean we expect the RBA to be looking at what’s next. Moving beyond that, here are three other signs pointing to the idea that the RBA could be open to more easing, apart from the fact that the economy is weak.

1 – The RBA has not kept pace with everyone else

To date, the RBA has only executed one third of the amount of easing via QE compared to other major central banks. Chart 3 shows that the RBA has announced policies that would buy about 10% of assets in GDP, whereas the asset purchase programs and funding schemes of other major countries have already exceeded our maximum, with some indicating they’d go as high as 30%. One could conclude that either the other countries have done too much or the RBA too little. Chart 2 hints at the answer: too little.

Chart 3 Central bank facilities* (programs announced since 3 March; % of GDP)

Source: *Includes asset purchase programs (including loan purchases) and term funding schemes; excludes short-term liquidity facilities. **Maximum size assumes no further bond purchase; includes the AOFM structured finance support fund.
https://www.rba.gov.au/publications/smp/2020/aug/pdf/statement-on-monetary-policy-2020-08.pdf

2 – The RBA has adjusted the way it is operating with QE

In recent times, the RBA has changed the way it executes its QE program. Looking at the timing of their purchases really bears this out. As an example, the RBA told us that they would buy bonds to keep the 3-year bond yield at 0.25% (dark blue line in Chart 4). This is essentially what they did during March and April 2020, buying bonds aggressively to stabilise yields. However, as soon as yield stabilised, they paused their purchases. From 6 May through to 5 August, the RBA bought zero bonds and indicated it was happy with the yield. However, from mid-August, the RBA started buying bonds again, even though the 3-year yield hadn’t changed.

Chart 4 RBA purchases and 3-year bond yields

What triggered this latest buying spree? My answer: Victoria entered stage 4 lockdown, knocking the chance of a v-shaped recovery back in time. This makes it look like the policy is about more than just keeping 3-year yields at 0.25%, instead adding the focus of buying assets to help ease conditions.

Chart 5 Stage 4 lockdown and QE

3 – The RBA told us they could ease more if they wanted

The last point to make is that the RBA has messaged that they could ease more if they wanted to. Here are some recent comments from Governor Lowe on additional QE and negative rates:

“We have not ruled out a separate bond buying program, or other adjustments to the mid-March package. But for the time being, the Board's view is that the best course of action is to continue with the current package.”

—Governor Lowe, Opening Statement to the House of Representative Standing Committee on Economics, 14th August, 2020

“RBA LOWE: HAVEN'T RULED OUT NEGATIVE INTEREST RATES”

—FXstreet.com, 14 August 2020

And more recently in their monetary policy statement. What are they considering?

“The Board will maintain highly accommodative settings as long as is required and continues to consider how further monetary measures could support the recovery.”

—Statement by Phillip Lowe, Governor: Monetary Policy Decision, 1st September, 2020

These comments lead us to the first big question.

Would the RBA use negative rates?

While the headline from fxstreet.com makes a compelling case for why we should be thinking negative rates are a possibility, the other side of that comment reminds us not to be too hasty in leaping to any conclusions:

“Lowe has said that negative interest rates are extraordinary unlikely in Australia”.

—FXstreet.com, 14 August 2020

Now you could run the argument that the RBA told us in late 2019 that QE was unlikely; yet six months later they introduced the policy. Meaning negative rates could be here sooner than we think.

While that is true, it does overlook the language that the RBA used to describe the outcomes. As an example, here are comments that Governor Lowe made in regard to both QE and negative rates in two separate speeches:

“While at this point it is unlikely that the Reserve Bank will need to employ unconventional monetary measures, the Reserve Bank Board considered it prudent to understand the issues involved and has studied the experience of other countries.”

—Governor Lowe, Questions on Notice, RBA, aph.gov.au, 19 September 2019

“The second observation is that negative interest rates in Australia are extraordinarily unlikely.”

—Governor Lowe, Unconventional Monetary Policy: Some Lessons From Overseas, 26 November 2019

QE was unlikely (but subsequently occurred) while negative rates are extraordinarily unlikely. Given the RBA is using that same language now to describe negative rate policy, it is likely some time away if they were going to try it.

But that is based on the current economic conditions. Let’s double back to the question we were trying to answer: would they try this policy if equity markets fall? My answer to this is “yes”, because an equity market decline would imply that the recovery in the economy is not going as planned and certain sectors were struggling. While there is some commentary from the RBA that negative rates have too many side effects, the research coming from offshore doesn’t always bear that out. Here are a two research papers to support this view, the first from the European Central Bank (ECB).

Negative rates and the transmission of monetary policy

“Overall, negative interest rates have supported economic activity and ultimately contributed to price stability.”

“Negative rates, in conjunction with the other policy measures, have contributed to the euro area expansion and supported inflation expectations.”

“Negative interest rates have had a broadly neutral impact on bank profitability so far, as their negative effect on net interest income has been offset by a positive effect on borrower creditworthiness.”

https://www.ecb.europa.eu/pub/economic-bulletin/articles/2020/html/ecb.ebart202003_02~4768be84e7.en.html#toc7

Going Negative at the zero lower bound: The effects of negative nominal interest rates

“Negative rates can stimulate the economy by lowering the rates that commercial banks charge on loans, but they can also erode bank profitability by squeezing deposit spreads.... I use bank-level data to calibrate the model and find that monetary policy in negative territory is between 60% and 90% as effective as in positive territory.”

https://www.frbsf.org/economic-research/files/wp2019-21.pdf

(Note that you can also find economists arguing the point the other way.)

This isn’t to say that the view on negative rates is all positive. Rather, when you have nothing left, you probably think it’s a more viable idea. So, if the equity market melted down and conditions were weak, I think we would find that the word ‘extraordinarily’ would disappear from the narrative.

Since the RBA recently told us that they are not expecting inflation to be back into their 2 – 3% band for the forecast horizon (which is a couple of years) and employment is forecasted to hit 10%, it would be nice to find a policy that helped the RBA hit its policy goals. The above ECB quote suggests that negative rates contributed to both growth and increased inflation expectations in the region. So it stands to reason that they should potentially be more than just considered.

As a segue into what this would mean for bond yields, here is the Australian interest rate curve vs peers, separated into two groups:

  1. Countries using negative rates (Japan + Europe—dark green & dark blue)
  2. Countries that have signalled they could go negative (NZ + UK—light blue & turquoise).

Chart 6 Australian yields vs negative countries

If the RBA was willing to go negative, the yield curve could easily drop around 100 points—pulling it in line with the UK. (If they wanted to adopt the more extreme European style, it could be closer to 150).

What happened in other countries that went negative?

This is an important consideration as it allows you to observe what rates do before and after negative policy is announced. Based on the way yields move in both QE and negative cash environments, we discovered that QE alone wasn’t enough to take rates negative. It is the threat (real or implied) from a central bank using negative interest rate policy that pushes them there.

Let’s circle back to the ECB research paper mentioned earlier, which supports this line of thinking:

“Negative interest rates remove the non-negativity restriction on current and future expected short-term rates and, therefore, monetary accommodation can propagate throughout the yield curve.

Before June 2014, the distribution of future expected short-term rates was effectively truncated at zero, as market participants were not assigning significant probability to future rates being negative. The introduction of negative policy rates allowed the constellation of rates to expand into negative territory. As the expansion of the interest rate distribution on the negative side effects not only current rates but also expected future rates, the stimulus also propagates to longer maturities.”

https://www.ecb.europa.eu/pub/economic-bulletin/articles/2020/html/ecb.ebart202003_02~4768be84e7.en.html#toc3

What the ECB is saying is that for rates to really go negative, the central bank must open up the possibility, otherwise the market won’t push bonds there themselves. Let’s take a look at some examples that illustrate this point.

The Japan experience
The first is Japan when rates went negative. Prior to taking rates negative, the 3-year bond yield flirted with zero, but it wasn’t until they moved the cash rate negative that it went consistently negative. The 10-year bond yield, however, never came close to negative prior to the policy, and has been crushed lower ever since.

It’s interesting to point out that the Bank of Japan (BOJ) only cut into negative territory once (I.e. they set the cash rate at -0.10% and haven’t moved it since). That red line represents the day of the move; a huge change in rates for an incremental 0.20% cash rate change.

Chart 7 Japan bond yields

A mirrored experience across Europe

The second is Germany. Prior to taking the deposit rate negative, Germany’s 3-year yield also flirted with negative, but just like Japan the 10-year yield was nowhere near it.

Once negative rates were introduced, the 10-year drops beneath zero and doesn’t look back.

Chart 8 ECB bond yields

It’s a similar story for Switzerland and Sweden, although Sweden doesn’t quite look as drastic, because Switzerland and the ECB got in a few months before them.

Chart 9 Swiss 10-year yield and negative policy

Chart 10 Sweden 10-year yield and negative policy

A more positive yield story

Alternatively, we can look at the yields of a few countries that have not yet used negative rates. Here you can see that if the central bank doesn’t hint at or use negative cash rates, then longer bonds don’t go negative.

Chart 11 shows the 10-year yields of Canada, UK and the US. Quantitative easing alone hasn’t been enough to get rates through zero over the past 10 years, usually stopping around 50bps. More recently, the UK has dropped through to 0.2%, which reflects the fact they have said they are looking into the policy.

Chart 11 Developed Market 10-year bond yields

This tells us that for bonds to offer protection in the realms of returns of +5% or more, it’s likely that the RBA will need to at least signal they are thinking about negative rate policy (and more likely use it). This brings us back to our first question:

  • Do we think the RBA could take rates negative (or is that completely off the table)?

In my view, I believe it’s a decent possibility if the situation worsens, but the RBA is not quite ready to admit it (hoping instead that we see a recovery in the economy). One trigger that could bring this forward would be the equity market falling 15+%, in which case there would be a duration trade left in bonds when it is needed, if the RBA is willing to act.

Having said all that, my view on rates ideas could be considered slightly more extreme than others, so let’s examine something a little ‘softer’.

What about if it’s only QE?

In terms of unconventional monetary policy, I believe there is a hierarchy on how the market views it; ranging from policies we are most comfortable with to those that the market view unfavourably:

  1. Zero interest rates
  2. Government bond purchases (traditional QE)
  3. Yield curve control (YCC)
  4. Corporate bond purchases (QE)
  5. Negative rates

My concept of ‘softer’, from an RBA perspective, is the RBA stepping up their quantitative easing program in the government bond space. As discussed earlier, the RBA has only executed one third of the QE compared to other developed markets.

If we look at those developed market peers that have done more QE, you can see our bond yields are still relatively high. If we exclude the countries with negative rates (and those who have proposed it) then we should be looking at the yields of Canada and the US as comparisons. In this instance, if the RBA were to turn up their QE policy, then we could see yields fall around 30 – 50bps depending on what part of the curve it is in.

Chart 12 Australian yields vs peers

Taking a more succinct look at the question of “Would the RBA use it?”, there is much stronger evidence that shows it is more open to this than negative rates. The RBA is already using the policy, has started buying bonds even as 3-year yields are on their target, offshore has done far more volume in bond purchases than Australia and economic conditions are weak.

More importantly though, a recent RBA research paper says that QE works, stating that ‘unconventional policies can stimulate economic activity through many of the same channels as conventional monetary policy’. Below comes from the conclusion on their recent findings:

“Although the cash rate is now at its ELB, alternative monetary policy tools are available to provide stimulus to the economy. Policies that lower government bond yields and household and business lending rates are effective in further reducing the unemployment rate and increasing inflation even though the cash rate is constrained by the ELB.”

——https://www.rba.gov.au/publications/bulletin/2020/sep/the-economic-effects-of-low-interest-rates-and-unconventional-monetary-policy.html

This finding shows that the RBA is obviously open to using more unconventional policies. Again, if the equity market were to fall significantly, I would run the same arguments that I did for negative rates—that this policy would be used. However, I would add the following caveats:

  1. This policy is far more likely to be used than negative rates, as it sits higher up my pyramid of what people (and the RBA) are comfortable with (and has already been hinted at)
  2. It may not take an equity market meltdown to get us there and in all likelihood probably does not require this catalyst for it to be implemented over the next six months.

The question sitting in my head relating to the QE purchases so far is: Why does the RBA believe they can do one third of the easing of all the other markets? Is our economy handling the COVID-19 shock better? Are we still the lucky country? Or have they just not done enough?

Evidence of defence—WW2, the deficit & yield curve control

The last factor to explore is yield curve control, which has already been pursued by a few central banks, including the RBA and BOJ. For the RBA, additional measures could include extending the yield curve control out the maturity curve. The evidence we have of these policies so far is that they work, and that is not just from the RBA’s recent policies.

To understand this one we need to go back a little further in time. One of the interesting comments that you will see at the moment is that countries are running the largest deficits since WW2. You can see that from the US context here:

U.S. budget deficit to hit record $3.3 trillion this year, CBO says

“Compared to size of the U.S. economy, this year’s fiscal deficit will be worst since World War II”

——https://www.marketwatch.com/story/u-s-budget-deficit-to-hit-record-3-3-trillion-this-year-cbo-says-11599070944

Here is how that deficit would look compared to the past 100 years.

Chart 13 US budget deficit

Source: https://www.thebalance.com/us-deficit-by-year-3306306

The numbers are significant! From an economic perspective, economists typically associate a larger deficit with higher rates (since the government will be issuing more debt, increasing supply). This leads most to think that this should bias rates higher. But if this is the biggest deficit since WW2, what happened to interest rates in WW2?

The answer: not a lot. You can see in Chart 14 that US rates actually didn’t really move from 1939 through to 1950.

Chart 14 US interest rates and WW2

Why? Yield curve control.

During this period, the US government set a yield target of 2.5% and offered to buy unlimited securities to protect it. Here is a comment from the New York Fed:

“By mid-1942 the Treasury yield curve was fixed for the duration of the war, anchored at the front end with a ⅜ percent bill rate and at the long end with a 2½ percent long-bond rate… The 2½ percent long bond rate was quickly accepted, widely supported, and never challenged”

“Private demand for Treasury securities shifted out the curve, driving yields on longer term securities below the fixed pattern of rates (Figure 8). From the end of 1942 to the end of 1945, 3- to 5-year note yields declined by 35 basis points (from 1.48 percent to 1.13 percent) and long-term bond yields declined by 16 basis points (from 2.49 percent to 2.33 percent).”

——https://www.newyorkfed.org/medialibrary/media/research/staff_reports/sr913.pdf

Based on Chart 14 and the comments from the NY Fed, the starting point is that yield curve control wasn’t challenged at the time.

More recently, Japan has run a yield curve control policy, targeting around 0% for 10-year yields. This was put into place in September 2016 and here is the 10-year yield since then:

Chart 15 Japan 10-year yield and yield curve control

It’s safe to say the yield has been “around zero”. This means these three experiences—the US in WW2, Japan since 2016 and Australia since March this year—point to yield curve control working. Why wouldn’t it? The central bank tells you the level they want and if you try to push it away they buy bonds until you stop.

Looking forward, if things deteriorate, the RBA could very easily say: “Here is where we want 10-year yields to trade, and we are committed to buying bonds to keep them there.” This makes it harder to forecast what it means for yields, since it will completely depend on the level that the RBA chooses. While it stands to reason that the RBA would set it lower than market levels to give the economy some benefit, they could just as easily set it where the yield is trading. Overall, this feels like it would be the lowest returning policy for them to use for bonds, but it has the added benefit of capping the downside as yields won’t blow off higher.

Finally, would the RBA use this policy in times of stress? I don’t see why it wouldn’t be considered; they are already doing it for 3-year bonds. Although I doubt it’s as meaningful as the other policies, so it’s probably more of a tool they use if the economy is stagnant and they are trying to give it some certainty, rather than shocking it back to life.

Throwing out some numbers (buyer beware)

Giving consideration to scenarios I have explored, we can estimate what it could mean for returns using the following scenarios:

  • Negative rates: Rates fall 50 – 100 points
  • More QE: Rates fall 30 – 50 points
  • YCC: Rates fall 0 – 30 points

The composite index currently has a yield of 0.81% and a duration of six years, so I will use the following for our estimates:

1 Year Return = Yield + (Duration x Change in Yield)

This would give the following hypothetical returns.

Table 1 Hypothetical returns under different scenarios

Note: if the RBA were to go harder on negative rates (like the ECB), then the upper estimate would be too low. Alternatively, if the RBA decided not to respond to a declining equity market, I would expect rates to sell off in response to the market reacting to a lack of support.

Conclusion

A few important points to remember:

  1. This analysis is prefaced on conditions deteriorating rather than what the most likely direction is for yields. It is not meant to breakdown what would be the catalyst for higher rates, since that would probably mean the economy is strong and equities are performing well, detracting from the question: Do bonds offer protection?
  2. All of these ideas require action from the RBA; whether it’s buying more bonds or taking rates lower. If I didn’t believe either of these policies where possible, it would change my confidence in rates at these levels.
  3. If conditions deteriorate, the RBA still has enough tools to take bonds lower. Offshore has demonstrated that there is a larger tool kit than we are currently using—negative rates, additional QE, and longer dated yield curve controls. The bigger question is really: would they use them?