HomeContributorsFundamental AnalysisIs the Economy ‘Plucked’?

Is the Economy ‘Plucked’?

Downturns are usually short, sharp and arrive at irregular intervals; expansions are longer and slower. The current slow upward grind in unemployment is a different phenomenon, but beware of nostalgia for 1980s solutions to current inflation.

Look at a graph of the unemployment rate for almost any western economy, and you will notice a few things. The big increases are sharp and sudden, while the subsequent recoveries are slow and last longer. The timing of those upward spikes is irregular, with no obvious pattern. The height of the spike in unemployment is a good predictor of how far the rate declines in the recovery that follows. By contrast, neither the length of an expansion phase of declining unemployment, nor the size of the decline predict the size of the next downturn.

This pattern is a far cry from the notion of a business cycle, where output and unemployment oscillate round some average or trend. There is something different about downturns. They are not just like expansions but a bit slower. The eminent economist Milton Friedman described this as a “plucking” model. The economy, he suggested, was a bit like a guitar string. Most of the time it stayed close to the fretboard. There is a physical maximum amount of production possible at any one time. Occasionally, though, the economy could be plucked away from that maximum, with output falling and unemployment rising sharply.

Remarkably, the mainstream macroeconomic literature that central banks rely on has engaged little with this feature of the data. Forecasting models used by the RBA and its peers do not produce this pattern. Some authors have looked at “plucking”-type models, though. Their analysis confirmed that downturns predict the size of the following expansion, but the reverse is not true. Expansions do not “die of old age” by building up imbalances. Australia went for 30 years without a proper downturn: it turns out that this is entirely possible provided nothing bad enough happens.

These authors further argue that the maximum production level during expansions is buffeted by small shocks, some temporary, others positive and permanent – think new technologies and ways of doing things, or a policy change that encourages workforce participation. Downturns, by contrast, are triggered by large shocks that are usually temporary and arrive at irregular intervals. Some argue that it is best to think of economies as having three phases: normal expansions, which can continue for long periods; shorter-lived downturns; and subsequent recovery phases where growth can outstrip normal rates and unemployment declines, but at a slower pace than it rose.

Some downturns do not end in full recovery, however. This observation links to another strand of research showing that the output losses associated with financial crises and wars are not fully recovered. The experience after the Global Financial Crisis shows this pattern in a number of advanced economies. Standard models would interpret this as slower growth in potential output, perhaps feeding the Australian tradition of hand-wringing about insufficient productivity or reform.

While the pattern in the data is clear, the causes of “plucking” are far from settled, with some authors seeing boom–bust dynamics as inherent in the economic system rather than driven by shocks. Some recent papers – including preliminary RBA work – look to certain frictions in the labour market for an explanation. These include search frictions, which arise because workers and firms sometimes require time to find a good match with each other. Another at least partial explanation is that workers resist absolute wage cuts but accept falls in purchasing power when dollar wages are constant in the face of positive inflation.

An objection to this line of research is the finding by some BIS researchers that the plucking pattern is clearer in countries with more flexible labour markets – mostly English-speaking ones. The literature focusing on labour market frictions also does not explain where the originating pluck comes from, only why the data have the dynamics they do when those shocks occur. Something must be causing the “pluck” shock, if indeed it is a shock. The BIS paper suggests one possibility: the authors find that output in the manufacturing and trade sectors shows a clearer pluck pattern than construction. Construction downturns tend to linger, perhaps because of their connections to indebtedness and financial crises.

If we take seriously the patterns in the data showing asymmetric downturns and the plucking pattern Friedman described, there are clear policy implications. Currently, the models underlying central bank forecasts are symmetric by construction. They therefore systematically understate spare capacity during expansions. They will find a positive output gap ahead of a downturn, even when there was no evidence of overheating at the time. This oversight risks policy mistakes. The asymmetric nature of downturns also means that the benefits of macroeconomic stabilisation policy are mostly about cushioning downturns and hastening recoveries. In normal times, the benefits will be small.

We also note that the RBA paper mentioned above hypothesised a short-run NAIRU and a long-run NAIRU that are both pushed around by actual unemployment in a path-dependent pattern. This is an argument for “holding on to the employment gains” that may have been missed by people whose mental model of the economy is fully symmetric, without any plucking.

The plucking model started from empirical observations about large downturns. The kind of slow grind upwards in unemployment seen in Australia in the 2010s and in the past couple of years does not fit this pattern. It is a qualitatively different phenomenon requiring other explanations, including perhaps the path-dependent drag upwards of the feasible unemployment rate by the actual unemployment rate.

It is these kinds of considerations that underlie our current forecast for slow growth and not a downturn, even though this week’s labour market data were concerning. A further 25 basis points or so of rate increases is unlikely to be the thing that causes a severe downturn. Typically, downturns are not just slower expansions but are qualitatively different phases that require a trigger.

The one thing that makes me nervous is that the plucking literature finds that every downturn is different. The famous opening line to Tolstoy’s Anna Karenina, “Happy families are all alike; every unhappy family is unhappy in its own way,” would seem to apply to economies as well. And in at least one case, a forceful tightening of monetary policy was the trigger for a downturn. The so-called “Volcker disinflation” brought US inflation under control in the early 1980s. That episode is widely regarded as a success, despite the output costs. In the current circumstances, some policymakers and commentators might be tempted to believe that the same medicine is needed; nobody wants inflation expectations to become entrenched, after all. However, as RBA Board member Ian Ross has noted, the institutions of the 1970s that produced wage–price spirals are no longer operative. So there is no guarantee that a similar approach nowadays would be necessary or as effective.

Watch out then, if the discourse turns to arguments that countries “need a recession” to get inflation under control. A Volcker-style policy could just end up plucking the economy.

Get daily and weekly ActionForex analysis and key releases on our Telegram channel.

Westpac Banking Corporation
Westpac Banking Corporationhttps://www.westpac.com.au/
Past performance is not a reliable indicator of future performance. The forecasts given above are predictive in character. Whilst every effort has been taken to ensure that the assumptions on which the forecasts are based are reasonable, the forecasts may be affected by incorrect assumptions or by known or unknown risks and uncertainties. The results ultimately achieved may differ substantially from these forecasts.

Latest Analysis

Learn Forex Trading