Good morning.
- Bonhoeffer on technology.
- The Democrats getting press, more here.
- Undermining public education by the Admin, a feature not a bug?
- Candidates and boxes.
- Wind bottles and power.
- Of metric, model, and stimulus.
- Church and culture.
- Trust and reputation.
- The medieval warm period … and a correlation (which is not causation … but that doesn’t mean it isn’t either).
- A long conversation on morals and war.
- Freedom of speech and the word ‘nigger’.
Of metric, model, and stimulus.
Not a bad discussion of what it means to estimate the impact of the stimulus. But I think what gets lost here is that the difficulty of estimating the effects of a policy is not an argument against the policy.
To use the caddy hypothetical, he knows the golfer is 150 yards way, he knows on average the golfer hits 150 yards with the 7-iron therefore suggesting the 7 is the rational policy. Suppose the golfer normally uses the 5 which only yields 50 yards and would have without the suggestion from the caddy. It’s sensible for the caddy to argue his suggestion brought the golfer 100 yards closer to the hole than he otherwise would have been.
The debate is given away when the author acknowledges that the range of multipliers goes from 0.4 to 1.5. In other words stimulus works in a slack economy. How much it works is another question. If the coherent arguments against stimulus carried real support, the range of multipliers would at least go down to 0 if not negative.
Boonton,
I’ve not examined how multipliers are used … but it seems likely a multiplier less than one is deleterious, i.e., that not pulling that dollar from the economy would be the normal 1.0 multiplier and that when the government actions do less than that, …. they’ve squandered resources.
If the dollar was ‘pulled’ from the economy then the multiplier would have to be negative. This was the so-called ‘Treasury View’ that Brad de Long often mocks from the Great Depression era. This view is only applicable in an economy functioning at full employment (plus a host of other questionable assumptions)
Undermining public education by the Admin, a feature not a bug?
I notice the right seems to have mostly missed the Admin’s rather vigerous push for perfromance based teacher compensation (much to the chagrin of unions). NY, for example, actually has a law against such compensation which it is itching to repeal because Federal grant money specifically requires schools be open to at least experimentation with it. Instead we have this nonesense which doesn’t actually impact any schools in a significantly relevant way (notice all the horror stories are hypotheticals, no actual examples are given of schools harassed for expelling ‘too many’ Latinos or having too many Blacks fail the AP Calculus test).
BTW, reading the article I’m a bit surprised that statistics was viewed as less challenging than calculus. My impression from tutoring was that statistics had an odd impact on people. People who had traditionally struggled in algebra and calculus would sometimes take to stats like a fish to water while people used to the solid foundation of equations and equal signs would find stats perplexing with its ‘confidence intervals’ and hypothesis testing that didn’t tell you the answer but only gave you a probability of being right in accepting one particular answer.
Boonton,
How can a dollar be pulled from anything but the economy? Where else is there to get dollars but the economy?
Do you have a link to a mathematically sound definition of this multiplier?
Let me see what I can find for you regarding a ‘mathematically sound definition of the multiplier’. I suspect any intro 101 textbook should do fine but the models used in the formal estimates are more sophisticated (which shouldn’t be a problem for you since you’re keen on mathematics).
In regards to ‘pulling a dollar out of the economy’. Let’s take a step back. The economy in a year is GDP or the sum of goods and services produced during a year. GDP is generated by a production function (http://en.wikipedia.org/wiki/Production_function). A simple one is Sowell’s Q = f (K,L). I’m leaving out the exponents and ‘A’ (a ‘know how’ factor sort of like technological knowledge or skill). Basically production happens by the economy combining capital (K, things like computers, machines, factories, snow plows, desks, copiers etc.) with labor.
You can use other functions. For example agriculturally minded economists would add ‘L’ for land but conceptually you can probably lump that into K or ignore it since land tends to stay constant. Basically what we need to know is that you produce GDP by combining its inputs. The more inputs you have the more GDP you have. Of course diminishing returns are in effect so if you have a lot of L adding more L will probably not boost GDP as much as adding more K but again we are keeping it simple here.
A deep recession is under utilization of the inputs of production. In other words, today we have about 10% or more of L totally unused, also called unemployment. We also have lots of K unused. Rows of vacant houses in the former real estate boom towns, office buildings unoccupied, power plants running at 90% capacity because electric demand is down etc. We tend to focus on L because unemployed K doesn’t generate lines of office copiers at the unemployment office giving interviews to CNN. Stimulus basically taps some L and K and uses it to produce GDP. It’s not ‘taking’ anything out of the economy unless currently employed L and K are leaving their jobs to work on stimulus jobs…but with 10%+ unemployment why wouldn’t those former slots be filled with unemployed L and K.
The primary criticism of stimulus uses a model where there is no unemployment. Over the long run this is a valid model because the economy does tend to put more or less everything and everyone to some type of use. If there’s no unemployment then stimulus must pull someone and something from current employment to ‘work’ on it (whether that stimulus is traditional things like road construction or more modern stimulus lilke people collecting unemployment paying their electric bill and stocking up on groceries at Wal-Mart). At that level stimulus does have a higher bar to jump over, basically its a question of whether the gov’t project generates more benefit than whatever private projects are displaced because of that. In the long run, ‘stimulus’ basically becomes gov’t spending and is evaluated on the merits of gov’t spending (I’m glossing over the Post-Keynesian school which holds that modern economies can operate in a state of long term underutilization requiring some degree of constant stimulus)
The problem is that while the zero unemployment model is useful for the long run and ‘works’ mathematically in some respects (explaining why the economy suddenly generates unemployment in a mathematically rigerous way is hard), it simply doesn’t fit reality. Despite what Fox News is telling you, no business today is laying off workers because they think ‘stimulus’ is bad.