I no longer believe in capitalism. At all.

[Jaguar]

I'll post about it after I walk home and then get pizza.

[Ben Kenobi]

I'll post about it after I walk home and then get pizza.

Cool!

[giblets]

There was a "Great Depression" before we had what we call the Great Depression.

http://en.wikipedia.org/wiki/Long_Depression

[Ben Kenobi]

This is shaping up to be longer. Even the 'Long depression' had an economic recovery by 1879 when the US went back on the Gold Standard.

[Sir Og]

As I said previously - I'd like to know more about this precise rule that you and Kuci would like to implement. I'm amenable to a different standard than Gold.

Read up on NGDPLT, NGDP futures prediction market and how servers can stay in basements and run 24/7 following specific rules.

[notyoueither]

There were lots of recessions under the gold standard. Like, really really bad ones.

I understand why you're afraid of discretionary monetary policy; we've given the Federal Reserve an enormous amount of power. I can understand why you want to tie them to a predictable "rule."

But you need to make sure what the right rule is. Kuci and I, in fact, want a robo-Fed that follows a precise rule, too! But the gold standard isn't a good rule to follow.

This guy calls himself a historian?

Can anyone answer the question of what happened to Europe's economy when the Spanish began bringing boat loads of bouillon in from the Americas?

Can anyone tell us what the European economy was like before that event?

[Jaguar]

NYE, that's a good example. I'd forgotten it.

BK: Kuci and I are with Sir Og, and various economists, that Nominal GDP should grow at a very steady rate, probably about 5%.

The reason for this is relatively simple - wages, we have observed for many years, are "sticky." Once someone gets used to a wage of $40,000, you can't cut it back to $39,000. It just doesn't happen. And not just that, but we rarely see employers that are willing to hire new workers at a lower wage than the incumbent workers at the same job. And compounding the stickiness of wages are union contracts, that are often negotiated in advance. The logic behind this, as well as the data, is overwhelming. But sometimes, people's work really does get less valuable. So what happens if you have a year of bad real growth - nobody's fault, just an unlucky year - and we all become a little bit poorer? What if an entry-level construction worker wage of $40,000 should be falling to $39,000 to make the market clear, but instead they have a scheduled automatic raise to $41,000? Well then, expect that employer not to hire new people for a while, even if an employee quits. You get a shortage of jobs in the sector, because the price of the labor is just a little bit too high. Once these job shortages start happening in many sectors at once, you get a lot of unemployed people. This is essentially how large nationwide bouts of unemployment happen.

Nominal GDP targeting fixes this. Nominal GDP is the sum of everyone's wages, essentially. A 5% growth target says that in the aggregate, all wages will grow by 5%. So the typical worker gets a 5% raise. But some of them are free to get 1% or 8%, and those sectors can continue hiring the right quantity of workers. 5% nominal income growth per year is the sort of predictable, non-discretionary target you want, but it's also something that solves the central "sticky wage" problem of macroeconomics. If people's natural nominal wages always grow at a steady rate, we don't get problems like the one I described above.

How do we implement it? We create a prediction market in nominal GDP, where the finance types can bet on what the final number from the Bureau of Economic Analysis will be. Then, if our robot sees that the number is too low, it expands the supply of dollars until the finance types agree that nominal GDP will be on target. If it sees that the number is too high, it will contract the supply of dollars until we're on target. It will relentlessly stay focused on its goal with 100% credibility, because of its simplistic programming and its clearly-defined goal: a steady growth in the price of labor.

[Kuciwalker]

How do we implement it? We create a prediction market in nominal GDP, where the finance types can bet on what the final number from the Bureau of Economic Analysis will be. Then, if our robot sees that the number is too low, it expands the supply of dollars until the finance types agree that nominal GDP will be on target. If it sees that the number is too high, it will contract the supply of dollars until we're on target. It will relentlessly stay focused on its goal with 100% credibility, because of its simplistic programming and its clearly-defined goal: a steady growth in the price of labor.

Note that this is the least important part of the plan. The plan could probably be implemented without any real loss of effectiveness by the Fed more-or-less as it is now. You pass a law telling the Fed to target the level of nominal GDP, and then you also tell them to do something called target the forecast. What that means is that they don't get discretion - they have to print the number of dollars that make the forecasts of their internal economic research groups predict that we hit the target.

That sounds a little complicated, but it's really the same thing as "point your car in the direction you want to go".

[Ben Kenobi]

This guy calls himself a historian?

Oh, there's lots of short and sharp ones. I wasn't sure just how long the longest recession was under the gold standard. I thought it was 4 years - but there were 2 that were 5.

Can anyone answer the question of what happened to Europe's economy when the Spanish began bringing boat loads of bouillon in from the Americas?

Well - Spain went to war with all of Europe and spent most of the gold that they brought in.

Can anyone tell us what the European economy was like before that event?

The motivation to go west was because the East was cut off.

[Ben Kenobi]

How do we implement it? We create a prediction market in nominal GDP, where the finance types can bet on what the final number from the Bureau of Economic Analysis will be. Then, if our robot sees that the number is too low, it expands the supply of dollars until the finance types agree that nominal GDP will be on target. If it sees that the number is too high, it will contract the supply of dollars until we're on target. It will relentlessly stay focused on its goal with 100% credibility, because of its simplistic programming and its clearly-defined goal: a steady growth in the price of labor.

Five is too high. Long run - you'll create unemployment as wages outpace production. 1 to 5, with average of 3 would be perfect. You probably don't want exactly the same number every year, and you probably don't want people to be able to predict what the number will be. And you probably don't want more than 2 5's in a row.

Why not have two rates - baseline at three and then have it -2, -1, 0, +1, +2 and cycle/ Even if it's off - you won't be off by much and you can correct - say every 5 years or so.

[Jaguar]

Five is too high. Long run - you'll create unemployment as wages outpace production.

You seem to be under the impression that the price of goods will remain the same, and because we don't have 5% real GDP growth, the wages would get too high as we pretend to make people richer than they are.

In actuality, the price of goods goes up when you increase the money supply, wages, and nominal GDP.

you probably don't want people to be able to predict what the number will be.

Yes, you do! That's the whole point. With a predictable rule - like "national nominal always increases at 5%" or "a dollar is always worth 1/1000 of an ounce of gold," businesspeople can stop wondering what what the Federal Reserve will do, and instead they'll get on with their lives and do actual business. The reason we prefer the former rule to the latter rule is that the latter keeps nominal wages stable and predictable - which is a lot more important than keeping the price of gold predictable.

We are able to keep one value, one price, predictable. Just one. Every other value will be less stable, and have to adjust around it through market means.

That price should be "everyone's combined incomes." Because the labor market is so important, so cumbersome, and so complex.

[Kuciwalker]

kentonio: here is Brad Delong, a very partisan Democratic economist, saying the same thing as Jaguar and I re: taxes:

http://www.economist.com/economics/by-invitation/guest-contributions/tax-luck-not-thrift%E2%80%94and-do-it-progressively

[Ben Kenobi]

You seem to be under the impression that the price of goods will remain the same, and because we don't have 5% real GDP growth, the wages would get too high as we pretend to make people richer than they are.

In actuality, the price of goods goes up when you increase the money supply, wages, and nominal GDP.

I want the growth of the money supply proximate to real economic growth. Either end is bad. I believe the real economic growth for the US for the forseeable future to be around 3 percent. I'm actually quite a bit more pessimistic on real economic growth longterm. My demographic models are downright nasty. Demand is dropping faster than wealth accumulation - nations are still poor and getting old.

The reason we prefer the former rule to the latter rule is that the latter keeps nominal wages stable and predictable - which is a lot more important than keeping the price of gold predictable.

The market in itself fluctuates. The money supply needs to fluctuate as well.

That price should be "everyone's combined incomes." Because the labor market is so important, so cumbersome, and so complex.

I like your idea and I'm seeing where it is going. I just don't like how high you've got the peg.