We'll soon get the announcement. Here are a few possibilities, computed by Thomson/Reuters, using citation analysis.
If I had to vote, it would be a Taylor/Woodford prize. For a sense of this work, see: link, link. While inflation targeting was invented in New Zealand based on the intuition of cleanliness in public administration, Taylor and Woodford had a lot to do with being able to think straight about it.
It's interesting to wonder how publicly visible citation data can be used to predict the Nobel prize outcome. One would want some kind of model which consumes citation data and comes up with an estimate of the Pr(Nobel prize). E.g. if one strong idea suffices (example: James Heckman) then it suggests certain people who will get through (example: Paul Romer). If they want you to build a broad literature (example: Robert Lucas) then that yields a different profile of those who will get through. Hmm, I can't come up with an example of another Robert Lucas. Such a modelling effort will yield insights on the usefulness of summary statistics of citation data such as the h-index and the g-index. As an example, it's easy to test which of the h index or the g index have superior predictive power in a model of predicting the Nobel prize.
For an analogy, it's easy to take firm data and fit an ordered probit model which predicts the credit rating. Once this is done, you're holding a black box which is able to look at firms, and come up with a real valued measure of credit risk. In similar fashion, once a model is estimated, which maps citation data and generates a real number from 0 to 1 (e.g. the probit latent variable), then it can be applied to a full database of all economists, and yield a measure of how Nobel-like each person is. This can be easily layered into Google scholar.
Most people in India think of Raghuram Rajan as being the first Indian-born economist who became the chief economist of the IMF. What is not widely appreciated is that in the class of economists of Indian origin who are younger than Amartya Sen, Raghu is the best set of papers, as measured by citations.