If I was stuck on a desert island, tasked with monitoring the health or otherwise of the global economy and allowed access to only one snippet of data a day, I know exactly which data series I’d choose: the yield on ten year US Treasury Bonds (i.e. the interest rate paid to people holding US government debt due in ten years’ time).
Now, I realise the above set of circumstances are unlikely to actually occur, but if they did I feel I would at least be mentally prepared.
Knowing what the ten year is yielding wouldn’t tell me everything about the health of the global economy, but it is a good place to start.
In many ways this is the key financial market variable — the so-called risk-free rate against which many other assets are priced. Is a ten year corporate bond paying 6% a year good value? It depends on the comparable returns available from investing in USTs (United States Treasury bonds). Is the US stock market a sensible investment if it trades at a price to earnings ratio of 12x and with a dividend yield of 3%? It might be, but again it depends on what the ‘risk-free’ alternative is yielding.
The outlook for the world economy at the moment (or at least for the advanced economies) is looking good. Certainly better than a few months ago, US and British growth is looking decent, even the previously-moribund Eurozone is showing signs of life. One of the bigger questions in macroeconomics though is: how will the US and global outlook be affected by rising interest rates?
The US Federal Reserve is widely expected to raise short term interest rates (Fed Funds) this year, that’ll be the first rate hike in almost a decade. The extent to which central banks raising short term interest rates (the ones they directly control) impact on the wider economy will really be determined by how bond markets react. The question will be: what do increasing short term interest rates mean for longer term ones?
The key variable to watch as the Fed begins raising rates, will be the US 10 year yield.
In theory, the future path of US interest rates should be priced into the ten year (i.e. the current yield of the ten year should take into account likely moves in short term interest rates). In reality this may be far from the case.
Usually, and in the text books, one would expect that a Fed tightening cycle (the US Federal Reserve raising rates from the cyclical low to the cyclical high over months or years) would be accompanied by rising bond yields.
In January this year though, despite widespread expectations of Fed rate hikes, the yield on the ten year actually fell. I’ve blogged before — at self-indulgent length — on why that might have happened. In February some of that fall was reversed.
In the past 20 years the Fed has embarked on three tightening cycles — in the mid 1990s, the late 1990s and the 2000s. The three cycles of rising short term interest rates met different responses from long term rates.
These three historical episodes give us three scenarios for 2015 and beyond.
The chart below shows US short term rates and the 10 year UST yield from 1993 to 2006.
The first cycle considered here began in 1994. 1994 was a, to use a not very technical term, messy year for global bonds with bond investors taking large losses (the price of a bond moves inversely to its yield — a rising yield means a falling bond price and vice versa).
Though in early 1994 Chairman Greenspan, in his usual Delphic style, had already been hinting at potential rate hikes for some time, the timing and magnitude of the subsequent monetary tightening came as a major surprise to analysts…
Over the course of the cycle almost all asset classes suffered losses, with global and Emerging Market equities hit the hardest.
As this BIS working paper from 1995 noted, the markets had not fully ‘priced in’ the extent to which the Fed was planning on raising rates. As the scale and pace of the tightening became apparent, bond rapidly repriced with the price falling and the yield rising. Long term interest rates rose sharply.
The ten year yield ended 1993 below 6% and a year later was closer to 8%.
The second cycle considered here runs from 1999 to 2000. This is as close to ‘textbook’ as the last twenty years of data comes. As short term rates rose, so did the 10 year yield. In fact the ten year ran slightly ahead of Fed Funds as the markets successfully ‘priced in’ the Fed’s likely course of action.
Finally there is the Fed tightening cycle that ran from 2004 to 2006. This is in many ways the odd one. Despite the Fed raising rates from just 1.0% to 5.25%, longer term interest rates barely budged.
This development may well have had serious consequences for the US economy. The mid-2000s were the time that the US housing bubble inflated, had the impact of rising short term rates passed through to longer term borrowing costs (and ultimately mortgage rates) then the bubble would have been much smaller.
In hindsight, the driver of this conundrum can be identified. The big factor at work was foreign buying of USTs. Rising foreign demand (especially from foreign central banks that wanted to control the value of their currencies against the US dollar), helped hold up the price of USTs and prevented a large rise in yields.
So — what to expect in 2015 and beyond?
What the Federal Reserve will be hoping for is a repeat of the 1999/2000 cycle in which the market understands its intentions and rising rates do not come as a shock to the market.
There is of course a risk of a 2000s ‘conundrum-redux’ scenario — a situation in which rising demand for USTs were to support the price despite Fed rate hikes. . If the Eurozone economy were to turn down again, if global risk appetite were to wain increasing demand for ‘safe haven’ assets or if global investors, disappointed by the ultra-low yields available on European and Japanese government bonds were to step up their buying of USTs.
That said, unlike in the noughties, if the Fed were faced by increasing UST demand holding up bond prices and leading to a monetary policy that is considered too loose (i.e. long term rates too low) it does a weapon available that it didn’t then.
One consequence of US quantitative easing is that the Fed currently holds more than two trillion dollars’ worth of government debt on its own balance sheet. It could, in theory, begin selling down these holdings to oversupply the UST bond market and push prices down and yields up.
The Fed will be hoping to avoid a 1994 outturn in which the bond market becomes turbulent and longer term rates rise sharply. Achieving that will require nuanced and clear communication.
There is no reason why the current tightening cycle can’t reassemble 1999 more than 1994. But there is one reason to worry. Liquidity in the UST market has been drying up. That is to say, the market has become less well functioning.
According to JP Morgan in 2014 it was possible to trade $280mn of USTs without moving the price. That has now fallen to $80mn.
Faced with tighter regulation and tougher rules on capital, the big banks that used to operate as market makers (buying and selling to facilitate trades and maintain market liquidity) are now less willing to.
This can lead to situations like those seen last October when the UST 10 year yield swung widely during the day.
The scary scenario for 2015 is a 1994 repeat coupled with a less liquid market. If the market comes to believe that is has under-priced the extent of Fed rate hikes and trades try to rapidly reposition themselves in a market where any sort of trading in size movers the price, then the outcome (to use the technical term again) would be messy.
Of course the Fed is well aware of this and will endeavor to avoid just such a scenario.
There will no doubt be more emphasis on communicating policy than in 1994. They will aim to avoid nasty surprises.
Hopefully they’ll succeed. But just in case they don’t, it is worth reading the 1995 BIS paper now.