During the subprime crisis some sectors had a productivity decline of about 15-20% (according to sna/bls data). Did the firms reinstall old machines from the attic for a few quarters? No. Will a firm immediately fire employees if turnover/value added goes down? No. Can the statistical office diffentiate between employees hard working at some times and picking in their noses at other times? No (for the most part). Labour hording can explain a great part of the productivity movement during the cycle (take for example a simple multiplicator-accelerator-model, assume some degree of labour hording and you will get tfp fluctuations with the "right" standard deviation and correlations). You don't need references for burning antic libraries...
Can someone please explain to me wtf a negative shock on technology is?
-
Any time the mix of inputs gets relativy more distorted from the optimal--either within firms, say, due to internal frictions, or across intermediate goods sectors, etc. Then measured aggregate TFP will fall--see Hseih and Klenow on their sectoral dispersion papers or Chari, Kehoe, McGratten on their business cycle accounting papers.
-
Y=f(A,K,L) is a model of the supply side of the economy people. For this reason, A tends to be explained from a production perspective only (vulcanoes, i-phones, oil shocks,...). But Y is a resultant of supply meeting demand so it's often driven more by demand factors, which are not part of the benchmark PF model. This can be seen particularly well if you take the cyclical and irregular component of A and correlate it with variables that proxy for demand (consumption, wages,..). They are highly correlated, more so than with variables that proxy for technology (R&D, I). Bottom line, negative (and positive) shocks to A often reflect demand and not technology or supply factors. The confusion comes from trying to explain A from the supply-side perspective (as it is a supply-side model), whereas short-run movements in Y (which A is derived from) are driven by demand factors.
-
Bad events that take the state of technology a step backwards:
- for example, you are a developing country, you piss of foreign overlords and they withdraw all their technical and technological assistance
- a war or revolution or fanatic religious takeover that gets many key researchers killed, universities with labs and server machines burned down
- some catastrophe wiping out a productive civilizationThey are other shocks which slow down technology's growth or evolution rate, that's a bit different and people here cited some:
- popular inventions that reduce productivity and training time of scientists or even the ability to learn by harming attention span for instance
- surge of anti-science sentiments and consequences like decreased funding for research
- industries which do not really further science but which get suddenly so well paying that they divert talented brains away from science
- etc... -
If z is negative in period t, how would you think about that in non-mathematic terms? Do machines suddenly decide to temporally become less productive?
It is a made-up convenience to help model "explain" reality without having to give up any of original wrong assumptions.
aka "economics research" -
Bad events that take the state of technology a step backwards:
- for example, you are a developing country, you piss of foreign overlords and they withdraw all their technical and technological assistance
- a war or revolution or fanatic religious takeover that gets many key researchers killed, universities with labs and server machines burned down
- some catastrophe wiping out a productive civilization
They are other shocks which slow down technology's growth or evolution rate, that's a bit different and people here cited some:
- popular inventions that reduce productivity and training time of scientists or even the ability to learn by harming attention span for instance
- surge of anti-science sentiments and consequences like decreased funding for research
- industries which do not really further science but which get suddenly so well paying that they divert talented brains away from science
- etc...Does anyone believe that these are important drivers of output fluctuations, though? Or better yet, has independent evidence not coming out of a NK DSGE style model? Explanations for TFP shocks appear to be not much more than a half-hearted attempt to make reality fit a fancy model, when really the model should try to describe reality. And of course, the same is true for the other gazillion black box shocks that people come up with. I’m happy to be convinced otherwise, it all just seems a bit useless to me.
-
Answer to can someone explain what a negative technology shock is:
Socratic method.
1. Read Diamond Mirrlees Optimal taxation and public production I: Production efficiency, PA Diamond, JA Mirrlees - The American economic review, 1971 - JSTOR
2. Ask yourself if you had data on a firm and anything whatsoever distorts the mix of intermediate inputs
i. actual taxes
ii. any adjustment cost
iii. any other frictions, search, congestion, clogs in delivery
will make the production plan inefficient.
3. Ask yourself what does it mean for the production plan to be inefficient?
4. What would an outside observer who just sees inputs and output with a given production function and a derived productivity A_jt find happened to productivity when the frictions get worse.
5. When you answer 1-4 yourself, you will rediscover what Diamond and Mirrlees taught us a half a century ago. They PROVED when we will observe negative productivity shocks--very precisely.
6 Generalize their arguments to distortions across sectors. Then mix in the Farhi et al idea of production chains and you will have a nice answer--and nice paper if you write it up properly. -
If the op refers to a decrease in statistically measured TFP: most likely these are just statistical artifacts.
How’s that?
Let's consider L:
A barber shop has 40 customers on day A, the employees are working at full capacity. On day B, only 20 customers come, the employees sweep the shop half their working hours or pick their noses with their fingers. Most likely, the employer will report to the Bureau of Statistics for both days that his employees worked 8 hours. Labor input, capital stock and wage income remain the same, profit is halved, wage and profit income equals Y, measured TFP goes down.Let's look at K:
The statistically measured K of the production function is not the number of scissors actually used during productive working hours, but the (deflated) monetary value of the total available capital stock (the value of all the scissors that are lying around in the shop). The latter does not change with working hours. Even if productive working hours are measured correctly, any decrease of it goes with the factor (1-a) into TFP.And it’s getting worse:
With a little algebra, the identity Y = W+P can be transformed into the following relationship, expressed in growth rates (displayed with “d”) (w and p are the relative factor incomes):dY = a * dw + (1-a) * dp + a * dL + (1-a) * dK
So, the measured TFP change equals the income share weighted change of the factor incomes. As P goes up in a boom und vice versa, factor shares change during the cycle, and as a consequence measured TFP. Remember that p is defined as P/K. K does not move that much during the cycle, and P is highly positively correlated with investment demand (due to accounting identities).
In the end, most likely, (investment) demand drives the cycle and statistically measured TFP fluctuations are nothing more than its statistical reflection.
RBC to the toilet.