Lesson 4 of 8
Shocks, and an efficient cycle
The claim that fluctuations are the optimal response to changes in technology, stated by its proponent, with the objections a critic made in the same issue.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- The claim is that a model with no monetary disturbance and no market failure, driven only by variation in technology, produces fluctuations like the ones observed.
- The consequence its author draws is that costly efforts at stabilisation are likely to be counterproductive, because the fluctuations are efficient responses.
- The objections published alongside it are about measurement: the parameters, the shocks, the absence of prices from the exercise, and the major recessions it does not explain.
This is the account furthest from everything else in the track, and it is worth reading precisely for that reason. A quarterly review published by a Federal Reserve bank in 1986 carried the statement and the sharpest reply together, which is unusually convenient.
One issue, both sides
The claim
- Fluctuations follow from variation in the rate of technological change.
- Theory predicts output deviations of five percent and more from trend.
- It predicts investment three or more times as volatile as output, consumption half.
- On this account, if the economy did not fluctuate, that would be the puzzle.
The objections
- The parameters are not supported by the microeconomic evidence claimed for them.
- There is no independent evidence of technology shocks of the required size.
- The analysis uses no price data, which is an odd way to test an economic model.
- It offers no account of what actually caused the major recessions.
The disagreement is not about the model's internal logic. It is about whether its inputs were measured.
The claim, as its proponent states it
Prescott's argument is that a competitive equilibrium model of growth, with households choosing between work and leisure and between consuming now and later, and with a stochastic technology parameter, generates fluctuations quantitatively like those the US economy displays.
The summary he gives is confident. Theory predicts what is observed: the size of the deviations from trend, the relative volatility of investment and consumption, and the serial correlation of the deviations. He states that if the economy did not display these business cycle phenomena there would be a puzzle.
He also draws the policy implication explicitly, and it is the part most often reported out of context: that economic fluctuations are optimal responses to uncertainty in the rate of technological change, and that costly efforts at stabilisation are likely to be counterproductive. He immediately qualifies it: this does not mean the average rate of technological change is optimal or independent of policy, and he argues attention should shift to what determines that average rate.
The objections, from the same pages
Summers wrote a reply printed in the same issue. His four objections are methodological and they are sharp.
The parameters are not, in his view, credibly measured. He takes one parameter Prescott identifies with a share of household time and argues the data on its average value do not support the figure used.
There is no independent microeconomic evidence for technology shocks of the size the model needs, and he asks what a negative technology shock would physically be. He also notes that measured productivity moves for reasons other than technology, including labour hoarding over the cycle.
The third objection is the one that lands hardest: the exercise is price-free. An analysis of consumption and production that never confronts price data is, he argues, not a test in the way economists usually mean.
The fourth is about episodes. The account offers no explanation of how any particular major downturn came about, including the 1930s and the oil price episodes of the 1970s.
Where it sits relative to this track
At the opposite corner. Here the cycle is efficient, money is absent from the mechanism, and there is nothing to correct. In the account this track teaches, the cycle is a distortion caused by credit, and the correction is the bust.
One connection is worth carrying forward. Garrison notices that the time-to-build assumption, introduced into this literature to explain the persistence of monetary effects, is a step back toward representing the time structure of production, and he treats it as an opening rather than a concession.
Check yourself
What drives fluctuations in this account?
- Credit expansion by banks
- Variation in the rate of technological change, with households substituting over time
- Shifts in liquidity preference
Show the answer
Answer: B.
The model has no monetary disturbance and no market failure, and the stochastic technology parameter is what produces the movement.What policy implication does Prescott draw?
- That stabilisation policy should be more aggressive
- That costly stabilisation efforts are likely counterproductive, while the average rate of technological change is worth attention
- That monetary policy should target the price level
Show the answer
Answer: B.
He states both halves, and the second half is usually dropped when the first is quoted.Which objection did Summers press hardest?
- That the model is too complicated
- That the analysis uses no price data at all
- That technology does not matter
Show the answer
Answer: B.
He calls it price-free economic analysis and argues that a theory not confronted with price data cannot be said to have been tested.Sources
- Prescott, 'Theory Ahead of Business Cycle Measurement', Federal Reserve Bank of Minneapolis Quarterly Review, volume 10, number 4, Fall 1986: the model, the quantitative predictions for output, investment and consumption, the statement that theory predicts what is observed, and the summary and policy implications section.
- Summers, 'Some Skeptical Observations on Real Business Cycle Theory', in the same issue: the four objections concerning parameters, the absence of independent evidence for technology shocks, the price-free character of the exercise, and the lack of an account of particular recessions. Prescott's response to Summers is printed in the same issue.
- Garrison, 'The Austrian Theory of the Business Cycle in the Light of Modern Macroeconomics', Review of Austrian Economics, volume three, section three, for the observation about time-to-build.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- This lesson reports one statement of the research programme and one reply to it, both from 1986. The literature has moved a long way since, and neither the later refinements nor the later criticisms are covered.
- The negative technology shock example restates an objection rather than reporting any measurement.