ONE of the ways in which America’s economy leads the world has been, of late, an unrivalled capacity for sending mixed messages. The past six months provide a case in point. The year opened with things looking pretty good: strong growth in late 2014 had led the IMF to project that GDP would rise by 3.1% in 2015; the Congressional Budget Office, America’s fiscal watchdog, expected a 3.4% expansion. In March and April, though, bad news built up, and in May official numbers confirmed that tumbling investment and exports meant that over the first quarter GDP had actually been falling at an annualised rate of 0.7%.

Then the good news came back. In early June data from the Bureau of Labour Statistics (BLS) showed hourly pay rising at an annualised rate of 3.3% in the first quarter and a 280,000 increase in employment in May, both far better numbers than economists had predicted.

This hot and cold pattern is becoming familiar. In 2014 predictions of a more modest 2.6% expansion were also dashed by a weak start to the year. Since growth returned in 2010 it has never beaten 2.5% over a whole year, and often fallen short. This has led some economists to worry that America will never get back to its pre-crisis 3% growth rate and may instead be stuck in a low-growth rut.

If they are right, it would be a huge problem well beyond America’s shores. The world economy is short of momentum: the big emerging markets, previously reliable growth engines, are struggling. Brazil and Russia face deep recessions, and China’s growth is slowing. The IMF predicts that, for the first time since 2007, there could be growth in every advanced economy this year, at an overall rate of 2.4%. But without American impetus the global recovery could come to a halt.

Pessimists cite persistent weakness in consumption and an edging up of the savings rate as signs that this much-needed American growth cannot be relied on. Yet the pessimists, like the optimists, keep having their narrative interrupted by indicators which just don’t fit.

On the basis of the year to date, the optimists have the better case. Poor performance in the first quarter can be ascribed to two particular shocks, a fall in the price of oil and a rise in the value of the dollar. Both now seem over, and the former, at least, unlikely to recur. With household debt much lower than it was and wages rising, the economy looks likely to be stronger than many expect. The worry for the next six months is less that growth will not return, more that concern about financial bubbles will encourage the Federal Reserve to raise rates before wage rises gain enough momentum. Real prospects for further growth could thus be dashed.

Excluding the plunge that started in autumn 2008, from which it mounted a partial recovery fairly quickly, the fall in the oil price from $104 a barrel in July 2014 to $47 in January was the largest six-monthly drop since 1986. Such a fall was reasonably expected to be good news when it comes to consumption—but not so good for the companies that make America the world’s largest oil producer.

Those companies reacted with impressive speed: by November 2014, just four months after oil prices started to slide, the number of active oil rigs had started to fall, too (see chart 1). A year ago the Eagle Ford and Permian basins in Texas boasted 763 active rigs, according to Baker Hughes, a consultancy.

By the end of May they had 342. As well as winding down existing wells, firms have stopped investing in new ones: government figures for investment in mines and oil and gas wells show it falling by $20 billion (more than 15%) in the first quarter of 2015.
.

So the industry’s contraction was fast and deep. But the countervailing surge in consumer spending never arrived. Petrol prices have fallen by 20% or more across the country; on June 1st they averaged $2.66 a gallon (€0.52 a litre). Produce the price of which depends a lot on transport costs, including fruit, vegetables and dairy products, has become correspondingly cheaper, which is part of the reason why the overall inflation rate has fallen from 1.7% in May 2014 to just 0.1% now. Nevertheless, consumption in the first quarter of 2015 grew by just 1.8%, much more slowly than it had in 2014. Rather than spend their cheap-oil dividends, people saved them. The household-saving rate rose from 4.7% to 5.6% between August 2014 and April 2015. Some analysts blame a bitterly cold February, which kept shoppers indoors.

The second shock was the value of the dollar, which rose in part because people expected further growth to be stimulated by a cheap-oil spurt in spending that never came about. That expectation, on top of a 3.6% annualised growth rate in the second half of 2014, made higher interest rates look imminent. American bonds thus took on a new allure, and as investors piled in the dollar rose by almost 9% in trade-weighted terms.

The rise, exacerbated by labour disputes at California ports, hurt exports even more than might have been expected. Adjusted for inflation they fell at an annualised rate of 7.6% in the first quarter. The drop in net trade (exports less imports) nudged up America’s current-account deficit, which now stands at $410 billion (2.4% of GDP), with the fourth quarter of 2014 showing a 30% deterioration compared with a year earlier. And by making foreign sales worth less back home, the strong dollar also had an effect on corporate profits, which fell by $125.5 billion, or 5.9%, between the last quarter of 2014 and the first quarter of 2015. But the worst of this seems over; the dollar’s appreciation has halted and, indeed, partially reversed—it has depreciated 2% since March,

Shocked, but stirring
 
The optimistic view is that, having absorbed the oil and dollar hits, there is now nothing between the economy and its deferred growth spurt. Yet there are niggling worries. February may have been cold, but March and April were less so—and yet consumption dropped in those months, too. And industrial production has fallen for five months in a row; between March and April it fell by an annualised 3%.

When numbers like these keep coming it is hard to believe the slowdown is temporary.

Hence the idea of “secular stagnation”. First set out in the 1930s by Alvin Hansen, it is a way of explaining depressions from which economies fail to recover in terms of a persistent mismatch between the supply of savings and the demand for investment. Reviving the idea in 2013, Larry Summers, a former treasury secretary, suggested that demand for investment had fallen because of technological advances that reduce the amount of capital it takes to start a firm.

At the same time, the supply of funds with which to invest has become plentiful, as a combination of ageing and inequality (old and rich people save more), as well as foreign bond-buyers, push savings higher. With a low enough interest rate this could be put right. But if, for an interest rate to be low enough, it needs to drop below zero, sorting things out may well be beyond the powers of the central bank. Thus interest rates can be at the same time both extremely low and too high. The damaging results include low growth, slack credit demand, weak investment and stubbornly high unemployment.

Although this analysis fits some economies—notably the euro zone and Japan—disturbingly well, it is at best only a partial description of America. There are three areas where things don’t look quite as bad as the secular stagnationists would predict. First, if interest rates had been consistently too high, capital expenditure would probably have stalled. But non-residential investment is up by 8% in real terms since its 2008 peak, and by 35% since its 2009 trough. As the government share of GDP shrank following the crash, investment took up the slack. It is true that investment fell in the first quarter of 2015, but that can be put down to the revenue-starved oil sector. Leave the extractive industries out and business investment actually rose by an annualised 1%.

Nor do credit markets suggest that interest rates are too high. American bosses are in a bullish mood. Far from weak demand for funds, borrowing is soaring. Companies are issuing debt at record rates, with $609 billion raised so far in 2015, up by $40 billion on a year ago according to Dealogic, a consultancy. Bank lending to business is strong too, up by 12% in the year to April, according to Federal Reserve data. As some of this money is spent on new offices, machines or software, this surge in credit should end up in some new investment.
.

A reassuring reduction in Americans’ debts up to the beginning of this year is a second reason for cheer. The latest data show the average household is in robust form, says Aneta Markowska of Société Générale, a bank. The ratio of mortgage debt to GDP has fallen below 80%, back to its 2002 level (see chart 2). Total household debt is around 107% of disposable income, with annual payments taking up less than 10% of disposable income, the lowest amount since at least 1980. (By comparison household debt is 136% of disposable income in Britain.) Household net worth (total assets minus liabilities) is at a record high in real terms, and is close to its pre-crisis peak as a share of GDP.
.

The third reason is the labour market. The rosy BLS data on pay increases and new jobs are part of a longer trend. Joblessness has fallen consistently since 2010. At 5.4%, unemployment is well below the post-1970 average. The trend shows little sign of slowing. Data on job openings and turnover collected by the BLS show there were on average 5m job openings in each of the first three months of 2015, the highest level since records began in 2000. A faster rise in the number of openings than the number of hires suggests competition for labour might be heating up (see chart 3). The market is fluid too. The number of workers leaving posts has increased, because more workers are quitting: a monthly average of 2.7m workers have left their jobs in 2015. This fact, together with strong data on hiring, suggest that these workers are moving on to better jobs.

Disunited states
 
There are still reasons to worry. The most recent data show that America’s labour productivity is stalling. The figures released by the BLS in early June showed workers’ output per hour had fallen by 3.1% between the last quarter of 2014 and the first of 2015. The drop was marked in the durables manufacturing sector, which fell by 3.3%; it meant that productivity is up by only 0.3% in the past year. That said, American productivity is still up by 11% since 2007, meaning that it is not yet as worrying as that of its European peers (in Britain output per hour fell over the same period).
 
If America is not suffering from secular stagnation, the competing explanation is that the country has yet to emerge fully from the financial crisis. James Sweeney of Credit Suisse, a bank, sees the problem in terms of a patchwork of localised depressions. A state-by-state analysis of GDP lends support to this idea. Many regions are growing well: in 14 states real-terms GDP rose by more than 10% between 2009 and 2013. Oregon and Texas expanded by more than 15%—North Dakota’s shale boom pushed output up by an astonishing 55%. At the other end of the scale are ten states, including Maine, Missouri and Nevada, hardly growing at all.

This analysis suggests that the aggregate figures are being held back by states and regions that are yet to recover from the first stage of the crisis: the period of negative equity and tight consumer credit that followed the 2008 housing crash. An analysis of housing data seems to confirm this: growth has been weakest in states which suffered the biggest price drops in 2006-10; strongest where prices tumbled least (see map). But things are turning. Mr Sweeney finds that even in the states that suffered the biggest busts, previously tight credit markets are starting to ease.
.

Differing views on America’s ailment—a long-term stagnation, or a balance-sheet recession that is easing—influence what economists think should happen to interest rates, which have been unchanged since the end of 2008. The IMF is dovish. In a review published on May 28th its economists recommended that unless there is new evidence of strong wage- or price-inflation, interest rates should not be raised until early 2016. The Bank for International Settlements, a central-banking body based in Basel, takes a different view. Its economists worry that low rates lead to asset-price bubbles, and suggest that it would be wise to start tightening as soon as possible before such bubbles are further inflated by cheap money.

Janet Yellen, who chairs the Fed, and her rate-setting colleagues are likely to take a middle path. Minutes of their meetings show they watch the labour market closely. Each new data point showing rising wages and tightening markets makes a change in September more likely.

At the same time, a further rise in the dollar based on the expectation of such a rise could self-defeatingly delay it by reducing growth prospects again. Knowing the question to be so finely balanced, analysts will pore over week-by-week data to try and divine Ms Yellen’s intentions.

Easy does it
 
After six years of cheap money, higher interest rates will weigh on borrowers. Prudent investors should keep an eye on America’s firms and equity markets. Despite modest growth over the past five years, companies’ share prices have shot up, with the S&P 500 index rising by 95%. Firms’ earnings have not justified the rise: data collected by Robert Shiller, an economist at Yale University, suggests equities are valued at around 20 times earnings, more than 30% above their long-run average; adjusted for the business cycle things look even worse. Some punters are borrowing to invest, with NYSE data showing margin debt—lending to invest in stocks—rising by 14% since January from $445 billion to $507 billion.
.


Whether or not this adds up to a bubble, it certainly looks like a market that could stall when interest rates rise and other assets, like bonds and deposits, become more tempting. And it is quite possible that firms’ earnings will soon fall. Staff costs are rising: a combination of higher hourly pay and lower productivity mean unit labour costs rose by 6.7% in the first quarter.

Companies’ bullish borrowing gives them more interest to pay. As earnings growth has slowed, the number of American firms with low credit ratings (BBB- or below) has risen, according to data from Moody’s, a credit-rating agency. If earnings do drop then the mix of high valuation and leveraged investors will sharpen the fall.

The idea that the pain which may be in store for the equity markets can be lessened if rates are raised sooner may seem to add to the case for tightening. History, though, warns against such reasoning. In August 1929, worried about overheating stockmarkets, the Fed raised rates despite signs that the underlying economy was none too strong. The move did not stop the subsequent crash; instead it helped precipitate the worst recession in American history. In the mid-2000s central banks tried the opposite, setting rates for the real economy and leaving markets to correct themselves. The results were equally bad.

This is not 1929. But the lesson is that since markets and the real economy can move at different speeds, central banking must be about more than interest rates. New “macroprudential” tools such as raising capital, liquidity and leverage requirements offer central banks ways to curb excesses in equity markets and sectors with too much borrowing; better that the Fed make use of them than that it tighten interest rates prematurely.

Interest rates should be set with an eye to the real, not financial, economy. Keeping them low will help households in the hardest-hit American regions overcome their lingering balance-sheet recession. It will also ensure that the wage pressure that builds up is not yet another blip.

Better, then, to wait. America seems not to be facing all-out stagnation and its households—if not, necessarily, all of its corporate sector—will be able to absorb higher interest rates when they come. But the time for that is not quite here.