Nineteen Fed officials, nineteen different futures

The Federal Reserve raised its benchmark rate to a range of 3.00 to 3.25 per cent at its September meeting, and the headline from its Summary of Economic Projections was a median forecast of 4.4 per cent for the funds rate at the end of 2022. The quarterly projections are meant to show a committee in agreement and to steady market expectations. They tend to record its disagreements instead. Behind the tidy median sits a dot plot that looks more like a scatter chart. Nineteen policymakers looked at the same economy and came back with nineteen different futures.

The spread of the dots is a measure of policy risk, and investors should treat it as one. In calmer times the dots cluster tightly, because officials broadly agree on the neutral rate and on how fast policy feeds through. Today's wide scatter shows that the committee has no shared model of how this bout of inflation will end. Some members think policy is entering restrictive territory that will cool prices quickly. Others fear a wage-price spiral that will need rates above 5 per cent.

Where the median points

The projections reach well beyond this year. The median official expects the funds rate to stand at 4.6 per cent by the end of 2023, with no cut before then, and sees unemployment rising to 4.4 per cent that year from 3.7 per cent now, which is the committee's way of admitting that it expects to cost people their jobs. September's increase was the third of 75 basis points in a row, and the median implies more than another full percentage point of tightening at the two meetings left this year.

The growth forecasts are as sobering. The median official now sees output growing by just 0.2 per cent over 2022 as a whole and by 1.2 per cent next year, down sharply from the June projections. The dots then fall to 3.9 per cent by the end of 2024 and 2.9 per cent a year later, so the committee is also pencilling in a long descent that depends on inflation cooperating.

Paying for disagreement

For bond investors, the scatter is a form of model risk that cannot be hedged away. Traders pricing interest-rate swaps or short-term Treasury futures are pricing a shifting coalition inside the committee room more than any stable policy rule. If the gap between the most dovish and most hawkish projections is two hundred basis points over the forecast period, the yield curve has to carry an extra premium for policy uncertainty.

That premium is highest at the short end of the curve. The two-year Treasury yield trades against the whole range of possible committee decisions, and the median dot is only one point in it. When voting members drift further apart, volatility rises, bid-ask spreads in the swap market widen and companies find it much more expensive to cap their floating-rate debt.

The market has already paid some of that bill. The two-year yield went above 4 per cent around the meeting, its highest since 2007, and the dollar extended a rally that has taken it to a twenty-year high against major currencies. Borrowers on floating rates, from leveraged-loan issuers to commercial property owners, feel each shift in the dots directly, because the cost of an interest-rate cap rises with the volatility of the path as well as its level.

The balance also shifts with the calendar. Only 12 of the 19 participants vote at any one meeting: the seven governors, the president of the New York Fed and four other regional presidents whose seats rotate each January. In January the votes pass from the presidents of Boston, Cleveland, Kansas City and St Louis to those of Chicago, Dallas, Minneapolis and Philadelphia. A committee whose median sits between two camps can be tilted by the rotation alone, before any new data arrive.

The old anchors have gone

The committee is split because the usual benchmarks no longer hold. Neither the non-accelerating inflation rate of unemployment (NAIRU) nor the natural rate of interest (r-star) works as a dependable anchor any more. Hawks argue that the neutral rate has moved higher for good, pushed up by de-globalisation and bigger budget deficits, as well as by spending on the energy transition. On that view policy has barely reached neutral and much more tightening lies ahead.

The committee's official estimate of that neutral rate has hardly moved. The median longer-run dot remains 2.5 per cent, which means the typical official believes policy is already well into restrictive territory at the current range. That estimate and the hawks' arguments cannot both be right, and the gap between them explains much of the scatter further out.

The dovish minority worries about the tightening already in the pipeline. Rate rises this fast have built up a large backlog of restraint that the economy has not yet felt. Monetary policy works with a lag of about eighteen months, so most of the effect of the 2022 increases is still to come. For these officials, raising rates hard on the strength of backward-looking monthly inflation reports risks a needless recession.

The dots have misled before

History supports treating the chart with suspicion. The committee began publishing it in January 2012, and in December 2015 the median implied four rate increases in 2016; the Fed delivered one. In March 2021 the median showed no rise at all through 2023, a view overtaken within a year by the fastest tightening in four decades. The strongest reply is that the chair, not the median, sets policy: the September decision was unanimous, and dissents have been rare in this cycle. That discipline holds while inflation keeps surprising upwards, and it will be tested once the data turn mixed and each camp can find evidence for its own view.

What tips the balance

The data that could move the camps are known. The monthly core inflation readings, above all in services outside housing, will decide whether the hawks' wage-price fears gain ground, while payroll growth and weekly jobless claims will show whether the unemployment forecast of 4.4 per cent is optimistic. The next set of dots arrives in December, and a narrower scatter then would matter more to bond volatility than any shift in the median.

The dot plot now shows how divided the Fed is, more than where it is going. The median dot is an arithmetic truce between two camps with different views of the economy, and investors who treat it as a plan will be caught out when the data next tip the balance. Over the coming year, the sensible preparation is for sudden changes of course whenever new figures expose how thin the committee's agreement is, with short-dated Treasuries and floating-rate borrowers bearing the brunt.