The first NGDP futures market is getting started based on the ideas of economist Scott Sumner. The idea is that the expected U.S. NGDP (nominal gross domestic product) is the single most important macroeconomic variable, and that having a futures (prediction) market will provide valuable information into this variable (Scott estimates that if it works, it will be worth hundreds of billions of dollars).
Unfortunately, due to US gambling laws (I think), the market will be based in New Zealand and U.S. citizens will not be allowed to participate.
A futures contract is one where you agree to buy a specific quantity of an asset today for a specific price, but you don’t pay until a specified time in the future.
If you predict it will have future value $100, and it only costs $10 now, it’s worth buying, hence there will be more demand, driving the price of the futures contract up. On the other hand, if it costs $100 today, but you expect it will cost $10 in the future, then the futures contract won’t be worth as much, driving the price down.
We still expect the price to be about as good of an approximation of the future value as you can get (as long as the volume is high) - if you have a better prediction, you can make money off it! So the price of the future will reflect the best aggregate prediction of the future value of the asset.
This is essentially the efficient-market hypothesis. This is the inspiration for idea futures, a.k.a prediction markets.
For NGDP futures, they would create contracts like this:
An example contract would pay $1 if the US GDP for Q1 2014 were greater than or equal to $17,250 Billion and less than $17,500 Billion, based upon the [initial estimate for quarterly nominal GDP from] BEA Table in Section 1 – Domestic Product and Income, Table 1.1.5, Line 1. We would establish similar contracts spanning the intervals $250 Billion above and below the example contract, with two open ended intervals beyond those for outcomes above or below the ranges.
The prices of these contracts now would reflect the market’s certainty that the future NGDP would be within that range.
So the price of the future will reflect the best aggregate prediction of the future value of the asset.
Well, technically speaking the price of the future will reflect the capital-weighted opinions of the market participants. That is not necessarily the “best aggregate prediction”—it could be, but there are no guarantees.
You don’t agree to “buy it today”, you “agree to buy” it today. Both exchange of payments and assets take place in the future
The price of a future isn’t determined by prediction any more than the asset is. The price of a future is given explicitly by the price of the underlying, suitably adjusted for cost of carry and cost of financing
The price of a future is given explicitly by the price of the underlying, suitably adjusted for cost of carry and cost of financing
First, that’s not true, technically speaking. The price of the future is whatever the market clears at. Arbitrage is a strong force that keeps the future and the underlying prices in a certain relationship, true, but only under certain (though common) conditions.
Second, here we are talking about NGDP futures and with them specifically there is no arbitrage against the underlying because the underlying is just an economic number that you cannot buy and warehouse. So in this particular case the price of the future is purely prediction-based.
It exists but it’s noisy. US Treasuries are more a bet on interest rates (i.e. monetary policy) than on actual growth; CDS is more about the possibility of a technical default caused by political brinkmanship than about the US genuinely running out of money. Based on a quick internet search, the S&P 500 is only very weakly correlated with GDP. Commodity prices are more about expectations for that commodity, foreign exchange rates are inseparable from the foreign country involved.
The first NGDP futures market is getting started based on the ideas of economist Scott Sumner. The idea is that the expected U.S. NGDP (nominal gross domestic product) is the single most important macroeconomic variable, and that having a futures (prediction) market will provide valuable information into this variable (Scott estimates that if it works, it will be worth hundreds of billions of dollars).
Unfortunately, due to US gambling laws (I think), the market will be based in New Zealand and U.S. citizens will not be allowed to participate.
To what extend do traditional finance markets provide an implicit prediction markets for future macroeconomic states?
A futures contract is one where you agree to buy a specific quantity of an asset today for a specific price, but you don’t pay until a specified time in the future.
If you predict it will have future value $100, and it only costs $10 now, it’s worth buying, hence there will be more demand, driving the price of the futures contract up. On the other hand, if it costs $100 today, but you expect it will cost $10 in the future, then the futures contract won’t be worth as much, driving the price down.
We still expect the price to be about as good of an approximation of the future value as you can get (as long as the volume is high) - if you have a better prediction, you can make money off it! So the price of the future will reflect the best aggregate prediction of the future value of the asset. This is essentially the efficient-market hypothesis. This is the inspiration for idea futures, a.k.a prediction markets.
For NGDP futures, they would create contracts like this:
The prices of these contracts now would reflect the market’s certainty that the future NGDP would be within that range.
Well, technically speaking the price of the future will reflect the capital-weighted opinions of the market participants. That is not necessarily the “best aggregate prediction”—it could be, but there are no guarantees.
A couple of points:
You don’t agree to “buy it today”, you “agree to buy” it today. Both exchange of payments and assets take place in the future
The price of a future isn’t determined by prediction any more than the asset is. The price of a future is given explicitly by the price of the underlying, suitably adjusted for cost of carry and cost of financing
First, that’s not true, technically speaking. The price of the future is whatever the market clears at. Arbitrage is a strong force that keeps the future and the underlying prices in a certain relationship, true, but only under certain (though common) conditions.
Second, here we are talking about NGDP futures and with them specifically there is no arbitrage against the underlying because the underlying is just an economic number that you cannot buy and warehouse. So in this particular case the price of the future is purely prediction-based.
It exists but it’s noisy. US Treasuries are more a bet on interest rates (i.e. monetary policy) than on actual growth; CDS is more about the possibility of a technical default caused by political brinkmanship than about the US genuinely running out of money. Based on a quick internet search, the S&P 500 is only very weakly correlated with GDP. Commodity prices are more about expectations for that commodity, foreign exchange rates are inseparable from the foreign country involved.