The EIA tracks production and consumption in terms of Total Liquids. Let's assume that we have production land and refinery land. Let's ignore refinery gains, transportation costs, energy used in energy production, etc.

Production Land produces 2 mbpd, and ships all of it to Refinery Land. Production Land consumes one mbpd of product, and Refinery Land consumes one mbpd of product.

So Refinery Land ships one mbpd of product to Production Land, and consumes one mbpd domestically.

On a net basis, Production Land shows a net total liquids export rate of one mbpd, and Refinery Land shows a net total liquids import rate of one mbpd.

Following is a slightly edited copy of a post I made on the 7/12 Drumbeat, regarding the ELM:

As Pitt pointed out, the ELM basically asks a hypothetical question for a hypothetical country (consuming 50% of production at peak production): What happens to net oil exports if production declines at 5% and consumption increases at 2.5%?

The answer is that net exports go to zero in 9 years, and only about 10% of remaining production would be exported. In any case, I suggest that you check out the ELM versus real data in Mexico post that Kkebab did. Also, the UK went from peak exports to zero exports in about six years.

In regard to economics, my premise is, and was, that once net exports started declining, oil prices would rise, generating increasing income for the exporters, even as their net exports fall. This would tend to have the effect, in short term at least, of increasing domestic demand in exporting countries. This is precisely what we saw in 2006. For example, the top five showed a 1.3% decline in production, a 5.5% increase in consumption and a 3.3% reduction in net exports (EIA, Total Liquids from 2005 to 2006).

IMO, almost everyone (Matt Simmons being a notable exception) in the Peak Oil community has been focused on the wrong thing--total production--when what counts is net exports.