Yield curve control has gone from more of an abstraction to a policy that many of the worldâ€™s most influential central banks are currently using or considering using.
In the US, yield curve control, often known as YCC, hasnâ€™t been employed since World War II.
Back then, the very high economic demands of WWII required capped funding rates to avoid US government debt â€“ and eventual servicing requirements â€“ from getting out of control.
From May 1942 to June 1947, short-term interest rates were held at 0.375 percent.
(Source: Board of Governors of the Federal Reserve System (US))
Intermediate yields included a 0.875 percent yield on 1-year bills, a 2 percent yield on 10-year notes, and a 2.25 percent yield on 16-year bonds. Yields of 2.5 percent were assigned to 31-year issues.
(Source: New York Federal Reserve)
The specific pattern of rates chosen was simply a matter of the financing rates that existed at the time they were implemented. As the country was still recovering from the economic malaise of the 1930s and Great Depression period, interest rates were already low.
There is no evidence of any attempt to identify market-clearing rates appropriate to a wartime economy. Nominal rates were kept fixed as a matter of policy irrespective of growth and inflation trends.
Under yield curve control policies, the central bank ensures low interest rates on borrowing many years into the future. The price of credit is set to a level that is close to the short-term interest rate, or cash rate, that the central bank is known for adjusting.
Yield curve control is often targeted at a certain maturity, such as the 10-year borrowing rate. Or it can be issued on a series of borrowing rates, such as debt issuances that span from three years to twenty years, for example.
In contemporary cases (e.g., Japan, Australia, the US), yield curve control has been targeted (or considered) at a single maturity in the three- to ten-year range. In other words, in additional to controlling the short-term interest rate, the central bank will also take over a longer term rate.
YCC vs. QE
Central banks can also influence interest rates through quantitative easing (QE), or the purchase of longer dated government securities (and sometimes non-government bonds, such as corporate debt and ETFs) to lower longer term interest rates.
QE is a secondary form of monetary policy when the adjustment of short-term interest rates is no longer effective.
The different between QE and YCC is that QE has no specific bond yield target in mind.
With QE, the central bank will commit to purchasing a certain amount of bonds or securities to support asset prices and that positive feed-through into the economy.
Yield curve control has a specific bond yield in mind. It will make the purchases it needs in order to effectively guarantee the yield rather than simply being an influence on them.
This ensures that large government deficits can be financed without interest rates going up.
QE will take a policy approach that commonly works by buying a certain amount of securities per month. YCC, on the other hand, will operate by coming into the market automatically as needed to buy or sell a certain amount of securities to maintain a specific yield.
Naturally, as debt issuance increases, investors are less likely to buy it. More debt supply can have an effect on creditworthiness.
Moreover, in terms of basic economics, more supply in relation to demand lowers prices. A lower price raises the yield. Price and yield vary inversely.
That said, the commitment needs to be credible. Buying an â€œinfiniteâ€ amount of bonds to guarantee an interest rate isnâ€™t free.
The pressure comes through either the interest rate side or the currency side. Since the interest rate is essentially pegged, that means itâ€™ll have to go through the currency.
If a country has reserve status, it can â€œprintâ€ money to buy the bonds more than a country that lacks it.
Reserve status is largely a function of how much borrowing is done in a currency internationally. If there is widespread use and borrowing of a currency, that helps support demand for it globally.
If thereâ€™s global demand for a currency, then issuing more of it isnâ€™t likely to lead to interest rate, currency, or balance of payments problems.
Under yield curve control, the money printing requirements are often significant in order to purchase enough bonds to meet the target as well as ensure enough liquidity to support the private sector and prices of risk assets.
Bonds that are effectively pegged would make the asset class very different from the ones that traders are accustomed to.
Historical vs. contemporary cases
Since the WWII period in the US, markets have grown accustomed to bonds responding to shifts in expected growth and inflation.
Periods of higher nominal growth expectations (nominal growth is the sum of real growth and inflation) have corresponded to higher bond yields and lower nominal growth periods have corresponded to lower bond yields.
Nominal growth needs to be higher than nominal interest rates otherwise debt will compound faster than the economy grows. Accordingly, you can see throughout time that nominal growth has usually exceeded short-term debt yields.
(Sources: BEA, Board of Governors)
The Bank of Japan has used YCC since 2016. Its cash rate was already at minus-10bps, with the 10-year yield targeted at zero, giving a steepness of just 10bps.
The Reserve Bank of Australia instituted its own form of YCC in March 2020 in response to the Covid-19 pandemic. In its case, it targeted 3-year yields and put them at close to zero.
Yields on 3-year Australian bonds
The Federal Reserve is considering a similar YCC program, which would also tentatively target 3-year yields.
Looking ahead, how long central banks maintain these policies will depend on how successfully their economies return to full employment and inflation targets and how tolerant central banks will be if inflation rises materially above targets.
Inflation may not typically be a problem in economic recoveries in developed markets, but if inflation trends up to 3 percent while employment is not at an adequate level, then it puts central banks in a bind.
Characteristics of yield curve control policies
The Fed during World War II and BOJ starting in 2016 set interest rates out more than a decade.
The RBA is doing yield curve control on a smaller scale, only targeting 3-year yields. The Fed is considering the same.
Of course, this can always be expanded into longer maturities if necessary.
Yield curve steepness
During World War II, the Fed targeted a relatively steep yield curve, with spreads between short-term and long-term interest rates at about two percent.
This helped bond investors enjoy steady returns and incentivized the private sector to help finance the government deficits.
On the other hand, a flat yield curve has been targeted in modern cases with the BOJ and RBAâ€™s programs (and the Fedâ€™s proposal).
During WWII, the Fed expanded its balance sheet, providing liquidity to the private sector, who in turn used some of this to purchase government bonds.
Australia and Japan have not been as aggressive in printing money, though have been willing to in order to meet their goals.
Naturally, the lower the funding rate provided, the more will need to be printed in order to buy enough debt to support these low yields.
Yield curve control policies are most effective when the fiscal arm of the government takes advantage of these low rates to finance large budget deficits and put the money in the economy directly.
This could be financing tax cuts or instituting something like â€œhelicopter moneyâ€ where the money is put directly into the hands of households with incentives to spend it.
When adjustments of short-term interest rates and asset purchases cease to work effectively in stimulating an economy it’s inevitable that the monetary and fiscal arms of the government must coordinate. Some call this “MMT“.
WWII saw huge fiscal deficits at around 25 percent of GDP. The response to the coronavirus also saw the largest deficits since the 1940s as the government sought to replace large amounts of lost income.
Yield curve control configurations: US WWII vs. contemporary forms
In each case â€“ the US in the 1940s and contemporary cases in Japan, Australia, and potentially the US â€“ interest rates had been pushed down about as far as they can go.
Nominal rates can be pushed down to zero or a little bit below. Once they reach those constraints private sector credit creation is difficult to stimulate much more. Creditorsâ€™ lack of returns dis-incentivize further lending.
Short-term interest rates as well as long-term interest rates are lower today than they were in the 1940s war years.
The US policy situation back then featured a fairly steep yield curve (200bps spread between short and long yields) and significant liquidity production at the front end (i.e., the Fed buying short-term bonds) to finance a massive fiscal deficit of some 25 percent of GDP during peak war spending in 1943.
(Source: U.S. Office of Management and Budget)
Japanâ€™s case, up until recently, was different from the Fedâ€™s WWII situation given that money printing and the fiscal deficit was nowhere near as extreme.
But Japan, in response to the economic damage from the Covid-19 pandemic, had to borrow more to stimulate and ramp up liquidity production for the BOJ to buy government bonds.
Australia needs to finance a very large government deficit (around 15 percent of GDP) with little money printing and only a target on 3-year bond yields.
Yield curve control as a hard rate setting policy on long-term yields
The Covid-19 pandemic stacked a lot more debt onto an economy that was already very indebted. To prevent debt servicing commitments from getting out of control, that means rates need to be held low.
QE can accomplish this, but has no specific rate target associated with it.
Adopting yield targets specifically asserts that an economy requires accommodative financial conditions.
It is borne out of a combination of limited stimulative capacity with short-term interest rates at zero, asset purchases already flowing, and large fiscal deficits.
In any country â€“ some sooner than others (particularly emerging markets) â€“ this will lead to an undesired rise in interest rates as more debt is supplied to the market.
The Fed has studied other central banksâ€™ uses of yield targets, most notably Japan (as it reached the zero lower bound decades ago). The desirability of the policy is mixed within the Fed, though itâ€™s practically inevitable that theyâ€™ll implement the policy.
Like Australia, hard yield caps on shorter duration bonds will likely be done first (e.g., ~3-year issuances). This will provide some steepness to the yield curve. If necessary, they can switch into longer duration securities with time.
Federal Reserve guidance and market pricing already imply an extended period of very low/zero interest rates. With the level of indebtedness relative to income so high, the idea of raising interest rates is far off.
For example, if total debt to income in an economy is 4x across all levels, that means even a one percentage point rise in interest rates would raise debt servicing requirements in the US economy by around $800 billion or about 4 percent of GDP.
If $800 billion goes away from investment and purchases of goods and services and toward debt repayment, thatâ€™s a knock on economic activity. Moreover, when interest rates are low, it lengthens the duration of asset prices.
Lower interest rates discount earnings at higher multiples, helping to raise asset prices.
If the risk-free interest rate is zero and the risk premium on equities is three percent, that means earnings multiples are 33x (the inverse of that three percent â€“ i.e., 1 divided by 0.03).
Raising rates by just one percent could knock a significant amount off asset prices. Hypothetically if a stock market index is producing ~100 in earnings, a 33x multiple produces an index value of 3,300.
If rates rise by one percent and the risk premium of three percent holds steady (1 + 3 = 4), then theyâ€™re discounted at that four percent, which is now a 25x multiple. At the same earnings, thatâ€™s 2,500, or about a 25 percent drop.
The Q4â€™18 tightening in US monetary policy was one such example where the Federal Reserve raised interest rates by more than was discounted in the curve, producing about a 20 percent drop in asset prices.
Either the central bank will keep interest rates low or they will run into inflation constraints that prevent them from holding yields as low as they effectively need to.
A front-end yield target would formalize the Fedâ€™s commitment to providing easy financial conditions. This would contain any rise in interest rates driven by rising new Treasury issuance.
This would help improve growth conditions (productivity plus growth in the labor force growth) or raise inflation in the real economy (demand for goods in excess of their supply beyond a certain acceptable rate of change).
Federal Reserve Chairman Jay Powell described the goals of the policy in a Senate hearing a few months into the Covid-19 pandemic and when it might be implemented (though he emphasized that no decision had been made at the time):
â€œThe sense of it is that if the market, if rates were to move up a lot for whatever reason, and we wanted to keep them low to keep monetary policy accommodative, you might think about using it, not on the whole curve, but on some part of the curve.â€
Fed governor Lael Brainard also echoed theÂ â€œyield curve control-likeâ€ tools in the US to reinforce forward guidance by asserting that:
â€œThere may come a time when it is helpful to reinforce the credibility of forward guidance and lessen the burden on the balance sheet with the addition of targets on the short-to-medium end of the yield curve.â€
Given the central banksâ€™ inability to meet their inflation targets in the past years, most will rather err on keeping monetary policy too easy. It’s better than the alternative of tightening prematurely and toppling over economies and financial markets that will tolerate little tightening without a commensurate rebound in activity.
Such a stance means that yields will likely rise much less and considerably slower than would normally be seen amidst a rebounding economy. One way to interpret this is that bond volatility and directional bond trades aren’t likely to produce much.
It doesn’t necessarily mean thatÂ yields would not be allowed to moveÂ from their current levels. The free market still has say so. Getting permanently out of the Covid-19 doldrums will help to a degree by secularly lifting yields (i.e., higher nominal growth expectations).
But it does mean that central banks would likely fight anything they perceive to be a premature tightening in financing conditions.
Active monetary tightening nonetheless remains far off. The short end of the curve will largely remain anchored in all major developed markets at or below zero.
In the next section, we consider what a yield curve control paradigm can mean for bond traders â€” especially in relation to what investors have lived through in recent decades where supply and demand have still been the predominant forces in the bond market.
Yield curve control policies can work to anchor interest rates through swings in nominal growth
Traders and investors are not accustomed to the idea of pegged yields outside short-term interest rates, or what they earn on cash.
Pegged interest rates, when and where they occur, make bond trading radically different.
Under such policies, central banks are not as sensitive to moderate swings in the cycle and donâ€™t change policy around movements in growth and inflation. Bond yields stay steady.
For traders and investors who use nominal bonds for diversification this has important implications. This source is basically lost if bonds do not help offset drops in equities, as had been the case for US markets from 1981 through 2020.
Depending on the steepness of the yield curve, nominal bonds could still remain attractive at longer durations, assuming they provide some yield.
Yield caps essentially take away the prospect of drops in their prices. Essentially longer terms bonds can become like a form of cash where a certain yield is received without much, if any, price risk.
This of course assumes that the yield cap commitment is credible and isnâ€™t pulled to allow for higher rates. If and when such policies do end, there should be ample forward guidance from the central bank.
Yield curve control drawbacks
Every policy has drawbacks.
If the Fed is monetizing the government’s deficits directly, then it may cause inflation expectations to rise and undermine the long-run goal of price stability.
This occurred the last time the Fed engaged in YCC in the 1940s and into the 1950s. The US government relented on YCC in June 1947 when inflation was 17 percent.
This eventually led to the 1951 Treasury-Fed accord, which separated government debt management from monetary policy.
YCC could also distort market signals. Government officials receive information from the bond and equity markets regarding economic conditions to help inform their decisions. Controlling a part of the curve would do away with a portion of those signals.
An additional challenge is determining when to exit from such policies. Matters like zero interest rates, QE, YCC, and so on, are supposed to be temporary departures from normal policy frameworks.
When the economy normalizes, it would be important to communicate the exit strategy from a YCC policy in a clear manner to avoid any destabilizing outcomes.
Like any policy, credibility is key.
A steep yield curve can make longer duration bonds attractive
The returns of bonds and the attraction of holding them under a yield curve control framework depends on the details of the policy.
The relative steepness of the yield curve is a big determinant.
In the 1940s example, low bond yields at the front end of the curve and slightly higher bond yields in a standard upward sloping yield curve was a positive for bond investors.
Despite the huge issuance of bonds, the carry associated with borrowing at near-zero cash rates and buying longer term yields one to two percent higher locked in guaranteed returns.
However, returns eventually eroded in real terms. Central bank easing to support the massive fiscal deficits eventually drove up inflation rates both during the war and after.
Longer term bonds were still nonetheless better to hold than cash, as getting that 1-2.5 percent is better than getting close to nothing.
Today, yield curves are much closer to zero throughout the developed world. They are also flatter.
Accordingly, there is less opportunity to lock in carry.
The nature of the inflationary pressures is different as well.
The deflationary nature of the pandemic from incomes and spending decreasing was added onto the various disinflationary pressures already in effect (e.g., high debt to income, aging demographics, labor offshoring to cheaper markets, technology advances and greater consumer price transparency).
The biggest determinant of whether we see an end to the secular disinflationary period lies in the future path of monetary and fiscal policy.
Monetary policy has had no issue inducing inflation in the financial economy (i.e., the supply and demand for money and credit) through low interest rates and buying financial assets directly.
It has not been as successful in getting that financial asset inflation to move into the real economy (i.e., the demand for goods and services).
The measures by central banks to counteract the credit contractions in 2008 and 2020 largely filled the gaps to avoid deflation and were not inflationary in themselves.
With Japanâ€™s experience with yield curve control since 2016, the yield curve has been much flatter.
Cash rates have been -10bps with 10-year yields held around zero.
There also isnâ€™t much available in the way for price appreciation (because yields are already at zero and canâ€™t be lowered much more) and there is very limited carry opportunity (because short-term rates are about equal to longer-term rates).
As a result, returns for market participants have been week. Since the adoption of YCC in Japan, owning 10-year Japanese government bonds (JGBs) have provided no returns in nominal terms.
At the same time, there has been little issuance from the fiscal side and inflation has been hovering at or a little above zero. As a result, returns in real terms have been only slightly negative.
Japan nominal rate bond returns (indexed to 2016)
Purchases of government bonds by the Bank of Japan
Yield curve controlâ€™s implications for non-bond asset classes
Lower rates are supportive of stocks, holding all else equal.
A commitment to keeping borrowing costs low helps companies make value-additive investments.
If lower rates also make cash and bonds unattractive, then equities look better in relative terms.
The commitment to buy bonds whenever they hit a certain yield (or at the high point of a range depending on the details of the policy) is bearish for the currency.
That doesnâ€™t mean itâ€™ll necessarily decline relative to whatever itâ€™s being borrowed against, but yield curve control is a policy that involves creating money to buy bonds.
Reserve currencies will be able to do this without seeing depreciation compared to those without reserve currencies.
The negative potential effects on the currency preclude some countries from easing monetary policy in such an aggressive way. This is especially true for countries that have limited borrowing and use of their currency internationally.
Gold is priced as a certain amount of currency per ounce. When currency is created in excess of the global gold stock, that tends to be a favorable tailwind for gold over the long term.
The dynamic isnâ€™t that gold is going up (i.e., becoming more useful), itâ€™s that the value of money is going down.
Gold generally benefits from a weaker US dollar. The USD is considered the worldâ€™s top reserve currency and gold is most responsive to real USD rates relative to any other currency.
But it broadly applies to all currencies. The greater the depreciatory pressure on a currency, the more gold will rise in relative terms.
In general, gold is an alternative when other currencies donâ€™t work as well.
Holding all else equal, the easing of monetary policy is bullish for gold denominated in that currency. The lowering of real interest rates decreases the return of interest-bearing currency and bonds, leading to the desire to find alternatives.
Yield curve control (YCC) operates similarly to a policy rate. It’s a way for the central bank to control interests rates along some portion of the yield curve.
YCC is different from what most market participants are used to.
Instead of a yield curve where the front end is set by the central bank and the remainder by supply and demand from various participants in the bond market, the central bank caps interest rates on one or more maturities further out.
A cap on yields means a floor on prices for targeted maturities.
If bond prices of targeted maturities remain above the floor, the central bank does nothing. The converse is also true. If bond yields are below the floor, the central bank doesn’t intervene.
If yields rise above the yield cap (or range, depending on the exact policy), the central bank will purchase the targeted maturity bonds. This raises their prices and decreases their yields.
The most prominent examples of YCC throughout history includeÂ Federal Reserve policy during and after World War II (1942 to 1947), the Bank of Japanâ€™s YCC policy adoption in 2016, and the Reserve Bank of Australiaâ€™s policy implemented in March 2020.