Offbeat Question Day: Stimulus

This is a question, probably mainly for the right, but the those readers on the left might offer their two cents. 

The $800b stimulus package has now been acknowledged by the President and pretty much everyone with their heads not in the sand to have been a waste. This criticism is especially strongly held on the right. However, what if the stimulus package instead of being useless road projects and expansions of bridges to nowheres (for example expanding the Byrd airport) had only one single project/point to which it was aimed. That is to say, all $800 billion was allocated to installing 40-50 new Gen-IV nuclear reactors throughout the country accompanied by ‘emergency’ executive orders designed to steamroll any and all environmental and green-activist objections. Furthermore these plants might have been be small and quickly installable and all fast-tracked to be on-line by, say, late October 2010. 

Here’s the question(s).

  1. Would that have changed your opinion of the stimulus package? 
  2. Would that have changed today’s economy for the better? 
  3. Would the election in three weeks from now be trending differently? 

I offer that the my answer might be yes (and with an “alas” on the last) to all three questions. What d’y'all think?

21 Responses to Offbeat Question Day: Stimulus

  1. Acknowledged by everyone to be a waste? AS in the economy would have been better without it or the economy would have been better with a hypothetically different stimulus program?

    As for your questions, no no no no. Why would you think allocating all funds into one type of activity would be better? During your education in mathematics were you never introduced to resource allocation programs and linear programming?


  2. Boonton,
    Let’s assume that the power plants are in the 25MW range … 50 -> 1.25TW additional power for the grid. No more “rolling blackouts” and the bottom drops out cost of electrical power. So, after the ‘mythical’ Keynesian stimulus boost we are left not just with repaired broken windows but plentiful power (non-CO2 at that) which will spur industry and development to use the same, but might encourage a general “tech” boost/emphasis. I could go on … but you could too.

    During your education in mathematics were you never introduced to resource allocation programs and linear programming?

    Huh? What branch of maths is that?

  3. Resource allocation problems? Linear programming? Basically they are problems of the form where you are trying to optimize some function (minimize total cost, max. profit). You are given a budget and constraints. One classic problem is construct a diet that minimizes cost but achieves 2000 calories a day and incorporates certain percentages of calories from various sources (fat, protein, sugar etc). Milton Friedman, I believe, found you can feed a person for a year for something like $300, but it would be quite unbearable from a taste perspective. Anyway this sort of thing comes up in managemant science courses but I imagine the formal branch of mathematics might be linear alegebra.

    Let’s break the issues with your proposal down:

    1. It’s an ‘all eggs in one basket’ approach which is almost always sub-optimal.

    2. In terms of stimulus I’d say its problematic for two reasons:

    a. It’s highly distortionary. While some of the jobs in building a power plant require ‘generic’ skills, many are highly specialized. It’s not enough to simply have, say, engineers but you need engineers with specialized experience and education. You’re going to generate super-full employment in those limited specialized areas while leaving many others untouched. After the building boom demand will fall dramatically leaving a boom bust in their pay which I think would be counter productive. Everyone else, though, gets little ot nothing.

    b. Even with ‘executive orders’ steamrolling fast track approvals, You’re still going to have major problems getting the money spent quickly. Even ‘shovel ready’ projects require multiple phases where one phase cannot be started until a previous phase has been completed. At best I think it would have taken a full year to two years before serious spending would actually have happened. In contrast a significant portion of the stimulus bill was immediate spending with a payroll tax cut, unemployment & food stamps extensions as well as intermediate with assistance to the states for Medicaid.

    3. In terms of energy policy its questionable IMO. ‘Rolling blackouts’ are more often caused by a poor transmission grid and lack of ‘swing’ supply (gas turbines that kick on to supply a spike in demand). Nuclear reactors by themselves do nothing to improve the grid and tend to provide ‘baseline’ supply which is not a problem. If I was spending $800B to improve the energy infrastructure of the US (never mind stimulus considerations which only are a minus to your argument), I’d look at upgrading transmission capacity and do a major R&D push in storage. Car makers are already pushing battery technology hard but the electric industry has an advantage in that they don’t need to be concerned about the weight of storage systems since they don’t have to drive them around. A major amount of cost in the system is due to powering on ‘swing’ generators to meet demand spikes and then turning them off when demand subsides. (I’ve been told that for gas turbines, the preferred source of swing supply, powering up and down is the major cause of wear and tear on the capital). Improved storage capacity could tap unused baseline generation (usually at night time as well as some from sources like wind) and tap it first during peak demand.

    4. ‘Resource allocation’ again. Let’s forget about stimulus and just focus on a hypothetical $800B fund to improve the grid in the US. I would say the most efficient way to allocate that would be across multiple initiatives. Say 20% to conservation (reducing home and business consumption thru relatively unnoticeable improvements in efficiency, not making people take colder showers or whatnot), 20% to improved grid infrastructure to make it easier to move power around to where it’s needed, 20% to R&D on storage, 20% to R&D on generation (fusion, solar, wind etc) and 20% on resolving nuclear’s waste issue. Provide nuclear with a waste solution and the market will build the plants as they are needed.

  4. might be of interest to you.

    Basically the US capacity factor is around 50%. What that means is that plants are actually generating only about 50% of the power they theoretically could. This is, again, because our demand for electric power isn’t consistent. For example, while a non-solar based plant could produce power 24 hours a day, we simply do not use power evenly over 24 hours a day. We drop our consumption dramatically at night. So plants have to be built to supply capacity during the day that is simply unused at night time.

    A major problem with your proposed alternative stimulus is that it would lower our capacity factor making our use of capital even less efficient. This cannot be said of most of the construction in the real stimulus program (and please try to keep in mind that actual construction projects were a minority of the stimulus bill). Most of the road resurfacing, bridge painting and probably even the ‘runway’ were put to use the moment the asphalt or paint was dry.

    I’ll give you this though, I think you actually might have produced an example of the Austrian concept of ‘malinvestment’. If you added a large amount of unused capacity to the market you’d certainly drop electric prices. But you’d also dramatically drop the rate of return for maintaining existing investment as well as new investment. Not only would new installations of wind and solar cease, so would a lot of maintaince in existing nuclear, gas, coal and other types of generating stations. So would a lot of efforts to boost conservation (what’s the point of an energy efficient fridge if the electric rate falls dramatically?) You’d depart from a market based allocation of capital here and move towards a centrally planned based version. I leave you the exercise of explaining why your usual devotion to Hayek and similiar thinkers would not lead you to think this is a less efficient state of affiars?

  5. Last factoid: seems to indicate (if I’m converting the units correctly) that last year we used about 3.7TWh of electricity. So increasing generation by 1.25TW would be about a 33% increase in power. But the year before we used 3.9 so we are clearly not suffering from a generation issue. We are leaving generation capacity on the table (and it’s not like we’ve lost a lot of plants in the last year) relative to last year as well as the fact that just about all generators are only tapped for a fraction of the power they could produce.

    I could see some arguments for this proposal, namely cheaper rates would put pressure on more polluting coal plants (but even worse pressure on smaller renewable plants). Increased baseline capacity could open the door to electric cars (say special night time meters that plug into your pluggable car and charge you lower rates, tapping otherwise unusued nuclear capacity). And increasing the differential in peak prices compared with off-peak prices opens up profit incentives for ‘battery firms’ to arbitrage (charge off the grid at night, sell in ‘spurts’ at peak times).

  6. Boonton,
    So that’s a 33% increase in power. Say you wanted 25-50% of the cars on the road to switch to electric. 11KW/hr is one gallon of gas and 9.8 million barrels of oil are consumed in cars each day. You need not just a little more power on the grid to move to electric. Solar and wind ain’t gonna cut it. My little 33% bump in capacity ain’t going to cut it either. But … that would be what we “might have done” in two years.

    You’re thinking small. I’m not offering a move to marginally change the energy economy here. The suggestion was to radically move it. Most of that $800b wouldn’t have been in construction, but systemization and development. If you kept ramping the power availability up to where we have (as suggested in that MIT document we read earlier this year to 50-60 TW of nuclear on the grid) that’s what we’d be looking for … and that moves us to a different level of power available. Which was the point of my “stimulus” notion. We are ultimately an energy based economy. Cheap energy can fuel a lot of growth.

  7. I think you’re missing the difference between capacity and generation. We are using about 50% of capacity right now. The massive nuclear or coal plant that keeps the lights going in the stores, businesses, homes and offices during the day is only being slightly tapped at night. The only time when there’s a problem with too little power is ultra peak hours when swing generators must be turned on. These are costly because they are expensive to run and they are not on all the time.

    If there was a dramatic increase in the number of electric cars there probably wouldn’t be much of an issue. People would mostly plug their cars in at night which would utilize a good amount of that unused capacity. While I understand you have a fetish for electric cars and for nuclear plants that’s not enough to base a public policy on. The grid doesn’t need new nuke plants as much as it needs a better grid to more easily move the power that is generated to the right spots and a better battery capacity to make use of generation capacity at off peak times.

    Don’t get me wrong, I’m not saying we have no needs anywhere for new plants, I’m saying that their return is less dramatic than you’re thinking.

  8. Boonton,
    I’m not against improvements in electrical distribution … although I’m guessing spreading lots of small power plants will lessen the stresses on electrical distribution systems.

    You need to look into comparisons of the energy requirements (consumption) by autos compared to the electrical power generation. You’ll find the former greater …. so … if you carbon fellas wanna move to more electrical cars and be fanciful about it you need to not just figure you’re going to consume off-peak power.

  9. Or just price carbon correctly and let the market decide what to ‘move to’. Sheeze, central planners are annoying.

  10. Boonton,
    I see. No stimulus. Well, I’m cool with that too. :)

  11. I think you’re conflating two different ideas here:

    1. Energy program….or more precisely IF we wanted to spend $800B on energy what would be the best way to do it? Your argument seems to put the cart before the hoarse, ramp up to supply 50% of vehicles being pluggable. If you really want 50% pluggable vehicles then. My reason for analyzing your proposal as just an energy program was to give it the benefit of the doubt. You can have a worthwhile proposal that is good regardless of the state of the economy that would nevertheless not make a good stimulus program.

    2. Stimulus programs are supposed to increase demand rapidly (or at least offset falling demand). By all accounts the $800B stimulus package was successful at that. Of course its nicer if a stimulus package gives you some lasting benefit that goes on beyond the recession. For example, the Hoover Dam continues to produce power and jobs many decades after its construction was providing relief from unemployment. Viewing WWII as a type of stimulus program, we are still reaping spill over benefits in nuclear power, aviation, medical advances and so on. But a stimulus program that just replaces demand without producing lasting investments is ok too. The idea is that the economy should produce closer to capacity rather than below capacity.

    So actually the bar is a bit lower for stimulus programs. An idea like a new highway that may not make sense during full employment may make sense in a deep recession when its stimulative effects are combined with its long term benefits. So here if the choice was nothing or your program I’d go with your program. But that’s not the choice you presented, you presented your idea as an alternative to the actual stimulus program. In that its defective for several reasons which I’ll summarize below:

    a. Too concentrated on one industry – that makes it distortionary
    b. Too concentrated on one aspect of one industry – even worse distortionary
    c. Even with ‘fast track’ approvals spending would still be much more spread out into the future. Stimulus is most effective at the deepest pits of the recession, less effective coming out and costly during full employment. You would likely push spending into the post-recovery zone where it becomes costly rather than stimulative.
    d. Because you’re probably going to bid up prices and wages for limited resources specialized to nuclear plant construction, you’re probably going to dull your stimulus’s multiplier effect. (Give someone a $40K job and you’ll increase spending. Make a $40K job into a $150K job and you increase saving) Better to pull 500,000 off the unemployment rolls into modest pay work than to bid up 5,000 salaries to bubble heights.

  12. Boonton,

    Because you’re probably going to bid up prices and wages for limited resources specialized to nuclear plant construction

    You mean like laying in reinforced concrete, electrical wiring, and plumbing. Yah. All very very specialized. Oops. Wait, those things aren’t specialized.

    “c” I stipulated that the plants would be going on line … this month or in the neighborhood. Not “spread out into the future.” Now you might contest the feasibility of that … but I don’t think you’ll back that up with any specifics.

    As for “concentrated” on one industry. You mean like the space program in the 60s, which concentrated “too much” on one high tech industry, but which I’d argue reaped high-tech benefits for lots of other industries all over. Hmmm.

  13. You mean like laying in reinforced concrete, electrical wiring, and plumbing. Yah. All very very specialized. Oops. Wait, those things aren’t specialized.

    No they aren’t, I’m thinking, though, that a small portion of the work does require highly specialized labor and possibly materials. Since you’re trying to rush through 50 plants at one time these resources will be stretched very thin and you’ll be bidding them up. I’ll grant you there will be more ‘mundane’ jobs that can be filled here relatively quickly. Beyond that you really don’t know what goes into making a nuclear power plant (and the numerous errors you made in advocating using a nuclear bomb in the oil spill fiasco should teach a bit of humility here).

    “c” I stipulated that the plants would be going on line … this month or in the neighborhood.

    I missed that, even assuming location sourcing and super streamlined approvals and variances takes zero time how are 50 plants going to be constructed in less than a month? You’re not very familiar with project planning are you?

    As for “concentrated” on one industry. You mean like the space program in the 60s, which concentrated “too much” on one high tech industry, but which I’d argue reaped high-tech benefits for lots of other industries all over.

    Spread out over several decades. The Apollo program, for example, was about 10 years or so but only cost $170B in 2009 dollars ( That’s about $17B a year which is not really much at all. If you’re talking about this as a stimulus program of about 3 years your rate of spending will need to be 15.5 times that of the Apollo program. If you’re talking about having it all done in the neighborhood of a ‘year’ you’re at 47 times the rate of spending.

  14. Boonton,

    I missed that, even assuming location sourcing and super streamlined approvals and variances takes zero time how are 50 plants going to be constructed in less than a month? You’re not very familiar with project planning are you?

    Hmm. And I missed the part where the stimulus package was proposed and approved this month.

    How long did it take to design and build the B-29? The first prototype (crashed) in late 1943. Operational wings were flying in numbers against Japan a year later. Seems to me 2 years is about right if you put an effort into it. The Wiki on the B-29 notes: “Thousands of subcontractors were employed” Hmm. Sounds “stimulating”, eh?

  15. gives 36 months from ground breaking to the reactor going critical. Streamlined or not you’re going to have to add IMO at least 6 months for site selection and planning. You may get some complicated issues in regards to spending. The program might generate some spending before it’s started simply from suppliers ramping up in preparation to bid on the contracts (for example, cement companies in the area might start investing in added capacity even if the actual spending won’t start for another year)…but as stimulus you’re really spreading the spending out pretty far which is the problem… stimulus its too spread out as investment its too hurried.

  16. Boonton,
    So … your argument against it is that you don’t think it’s possible. Sort of like it being impossible to rescue miners from that deep in the earth, I think. When a project has a single focused objective like this and the resources to back it up things thought impossible can be done. Kinda like the “we don’t do this because it is easy, but because it is hard.” But perhaps we aren’t that people any more. At least you (and the left if you are representative) don’t think we are.

  17. yawn, it’s quite possible to rescue miners in a mine, it’s also possible to build 50 nuke plants. That doesn’t make your proposal better than the real life package that passed. and you’ve presented no evidence to the contrary other than your fetish for electric cars and nuclear plants.

  18. Boonton,
    OK. So … if it’s possible … then let’s compare ‘your’ stimulus to my suggestion. Both pump similar quantities of money into the economy, filling that “pool” the same. Yours however leaves no structural changes in its wake, just lots of roads repaved well before they needed and parks a little prettier and so on. Mine. Cheaper power and lots of possibilities for industrial and domestic consumption expansion.

    Recall, the difference between conventional and nuclear plants is that the former has higher operating costs the latter higher capital costs (or more simply fuel is the primary cost for the former not the latter). Once the plants are in … future energy gets really cheap … and having a system of putting more plants in quickly and more cheaply now is also a benefit.

  19. I didn’t say your proposal hade no rate of return. Clearly it beats no stimulus at all. In terms of the actual stimulus bill, though, it’s inferior for the reasons I gave ( I don’t buy that you can spend the money as fast as the actual bill did). In terms of long term investment it suffers from too many eggs in one basket…..or more formally diminishing marginal returns. $50B in new nuke plants yields X in present value returns. $100B does not yield 2X but <2X. This is why an optimal resource allocation solution would spread over multiple projects rather than concentrating on one big thing.

  20. Also you haven’t really confronted the capacity versus generation problem. Take paving roads. Yes it’s small, yes it’s nothing grand. It’s no Great Pyramid or Collesuem(sp) or Great Wall. Nonetheless, poorly maintained roads cost something like $250 a year to drivers in excess repairs and maintaince cost. If the road repaving project reduces that by $50 per person that’s value realized right now as well as additional resources freed up for investments in the future (such as private building of nuclear power plants funded either thru slightly higher rates now OR individuals tossing $50 or so into their 401K’s which gets transmitted to bond and stock offerings).

    Right now, though, we are only tapping a bit less than 50% of our generating capacity. You would have us do a rush job right now to add a huge amount of added capacity. The returns you’re projecting are likely to be a lot lower than you think. In the immediate future a lot of those plants will sit dark, being tapped only for a fraction of what they could produce.

    Leaving stimulus aside, wouldn’t it make sense to first optimize the grid as it stands today and then work on adding capacity as we get to the point when we actually start using existng capacity more fully? If an onslaught of electric cars taps that capacity so much the better but right now if you made a list of pressing issues lack of generation capacity doesn’t seem to be a major one.

    Now the fact is that demand pulls supply. If you increase electric cars gradually you will increase the private returns on nuclear plants so you can still get your nuclear plant fix. Likewise if you boost the ability of the grid to capture and store off peak power and use it during peak times you also increase the potential returns on nuclear plants since they are primarily designed to satisfy ‘baseline’ demand, not swing demand. Making it so they can ‘swing’ a bit more will make them more valuable.

  21. Interesting, looking at

    According to them, it costs about $16,000 to equip a home with a 14.4Kilowhat hour solar array. At $800B you can so equip 50 million homes. You eliminate the grid problem because the added generating capacity is used directly at its source rather than having to make its way through the grid. You have thousand of unemployed home construction workers who could be tapped to do the installations. The added capacity will last for decades (30 years is usually given as a ‘useful life’). You have zero waste problem, zero location problem. You also don’t get as much capacity utilization problem as solar production automatically peaks during most periods of peak demand and there’s no cost involved in ‘spooling up or down’ the generator.

    One objection could be that everyone who isn’t one of those 50M homeowners won’t get the free power from the panels. But you could structure the program with a lend lease structure where the homeowners would slowly buy the panels over time, initially though they would be owned by the electric company or the gov’t and the power they produce sold to the homeowner at prevailing market rates minus a portion for using their roofs thereby decreasing the grid demand and lowering prices across the board.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>