Anniversary: Our first year, graded

This column began a year ago on a simple premise, that the steepest central bank tightening in forty years would sooner or later collide with private balance sheets. The federal funds rate now sits between 5.25 and 5.50 per cent, against 2.25 to 2.50 per cent when the column started. The consensus that expected an immediate recession and early rate cuts has been proved wrong by an economy that refused to buckle.

Where the premise held

The collision came quickly, though rarely where forecasters were looking. Within weeks of the first column, Britain's mini-budget sent gilt yields soaring and exposed pension funds running liability-driven investment strategies, forcing the Bank of England to buy bonds to halt a fire sale. In March, Silicon Valley Bank failed after losses on long-dated securities met a run by uninsured depositors, and days later Credit Suisse was pushed into the arms of UBS. Each casualty had assumed that low rates would last and discovered that duration risk was real.

Inflation, the reason for the tightening, has behaved better than the balance sheets. American consumer prices were rising at 9.1 per cent a year in June 2022; by July this year the rate had fallen to 3.2 per cent, helped by lower energy prices and the healing of supply chains. Much of that decline came from goods and fuel, which the Fed does not directly control. Core inflation has fallen more slowly, and services prices tied to wages remain the part of the problem that higher rates are meant to solve.

The tightening was global. The European Central Bank lifted its deposit rate from below zero to 3.75 per cent, and the Bank of England has taken Bank Rate to 5.25 per cent. In each case the first casualties were institutions that had borrowed short or promised fixed returns on the assumption that money would stay cheap. The British pension funds and the American regional banks were running different businesses, but they were exposed to the same mistake.

A year of damage

Avoiding recession is a different thing from avoiding the consequences. In twelve months the yield curve has inverted sharply, three large regional banks have gone into receivership, and the cost of servicing government debt has doubled. Legacy cash buffers and fixed-rate hedges absorbed the first shock. The harder test comes as those protections run out.

The gap between two-year and ten-year Treasury yields went beyond a full percentage point, the deepest inversion since the early 1980s. Signature Bank followed Silicon Valley Bank into receivership in March, and First Republic was seized and sold to JPMorgan Chase in May. The Fed's emergency facility, which lends to banks against securities valued at par, stopped the runs but left the losses on balance sheets. Fitch's downgrade of America's credit rating in August added a fiscal chapter to the same story.

Government feels the arithmetic faster than households do. A large share of federal debt sits in Treasury bills that reprice within months, and the Treasury has borrowed heavily since the debt-ceiling deal in June, announcing bigger auctions of longer-dated bonds in August. A homeowner with a fixed mortgage can ignore the Fed for years, but the government refinances continuously, which is why its interest bill has responded to higher rates faster than almost any private borrower's.

Where the column was early

The column underestimated how long borrowers could hold out. Millions of American homeowners locked in 30-year fixed mortgages at rates below 3 per cent in 2020 and 2021, so higher rates have hit new buyers far harder than existing ones. Companies lengthened the maturity of their debt over the same period, and households carried pandemic savings into 2022, which kept spending going even as real wages fell. Monetary policy has worked more slowly than usual because so much debt was fixed before the tightening began.

The consensus forecasters erred in the other direction, and for a related reason. They assumed that the historical link between rate rises and recessions would hold on the usual timetable, so they expected job losses by the spring of this year. Instead unemployment stayed near 3.5 per cent for most of the year, rising only to 3.8 per cent in August, largely because more people came back to look for work. The column's premise has held, but its timing was early, and early in markets is expensive.

The 2006 parallel

The last time the Fed held its policy rate at 5.25 per cent was from mid-2006, after two years of steady increases. For more than a year the economy seemed to cope. House prices had started to fall, but unemployment stayed low and share prices rose to new highs in October 2007. The damage emerged slowly, as adjustable-rate mortgages reset and the securities built on them lost value, and it arrived all at once in 2008. The lesson from that episode is that the lag between tight money and broken balance sheets can be long, and that calm in the interval says little about what follows.

The parallel is imperfect. Banks today hold more capital, and household debt is mostly fixed-rate rather than adjustable. The vulnerable borrowers are elsewhere, in commercial property and in private credit funds that have lent to mid-sized companies at floating rates. Those borrowers are smaller than the subprime mortgage market was, but their lenders are concentrated among the same regional banks and lightly regulated funds that are least equipped to absorb losses.

The maturity wall

Those defences are wearing thin. Mortgage rates have climbed above 7 per cent, the highest in more than two decades, and the ten-year Treasury yield reached its highest level since 2007 in August. Commercial property owners face the sharpest squeeze, because office loans typically run for five to ten years and must be refinanced at far higher rates against buildings now worth less than when they were bought. Smaller banks hold a large share of that debt, which ties the next phase back to the one that broke in March.

The strongest counterargument is that the economy has already shown it can live with these rates, and that inflation may fall fast enough for the Fed to cut before most debt comes due. That is possible. But markets expect policy to stay tight well into next year, and every quarter at current rates brings another batch of cheap loans to maturity. The indicators to watch are the Fed's surveys of bank lending standards, delinquencies on commercial property loans and the federal government's interest bill.

Private balance sheets have held up better than most expected, but the refinancing reckoning has only been postponed, and it will arrive as cheap fixed-rate debt matures. The first losses are most likely to surface over the coming year in office loans held by smaller banks, well before the broader economy shows the strain.