I use the UNFCCC GHG data for sector and national GHG estimates. Download the CRF reports for the countries you are interested in to get the best GHG stats for power sector.  Download the CRF zip file here. 

To see direct GHGs from electricity production, as well as upstream fossil fuel mining and refining emissions, expand the .xls file for the year(s) you are interested in and go to Table1.A(a)s1.  This is where you get the best electricity sector, coal mining and petroleum product refining GHG statistics.

To see nation-specific GHG emission factors for well-head/mine-mouth and processing GHGs for liquid and gaseous fuels, not including crude extraction or finished fuel transport emissions but including heavy oil upgrading GHGs (which are blended into the refining stats), look at Table1.A(a)s2 and Table1.A(b). 

So far, you have looked at energy consumption GHGs only and have not yet seen crude fuel extraction GHG estimates. Section 1.AA.2.F in Table1.A(a)s2 gives you the upstream fugitive GHG factor for each of the fossil fuels. Then Table1.A(b) gives you an estimate of the combined extraction through processing/refining (including fugitive) GHG factors for domestically produced fossil fuels. The GHG factors in Table 1.A(b) usually differ from one nation to the next. But it is going to be rare that the upstream GHG factor for imports is actually the same as it is for domestic production. 

The Table A(b) GHG factor is usually a pretty good estimate of the upstream emissions factor for the domestic production for the country you are looking at, and not appropriately applied to imports. But it is currently the case that both the Canadian and US CRF reports use a common North American GHG factor for natural gas. This factor, therefore, understates the upstream GHGs associated with Canadian natural gas consumption while it overstates the US upstream GHG factor. So if you use the CRF factor to estimate the full fuel cycle GHGs for Canadian electricity produced from natural gas, you will end up with an underestimate.  For most analyses, however, this factor will serve you well enough.
 
For a little snapshot of the differences between Canadian and US natural gas CO2 (not GHG) factors, go here. Select "Indicators", "CO2 emissions" then (1) CO2 from the consumption of natural gas and (2) CO2 from the flaring of natural gas. Then select "Natural Gas", then (1) consumption and (2) production.

They divide consumption emissions by consumption volumes, and flaring emissions by production volumes. You will see that this dataset puts the Canadian CO2 emission rate per unit of dry gas consumed at 1.5% higher than the US rate, and Canadian CO2 emissions from flaring per unit of dry natural gas produced at 122% the CO2 flaring discharge rate for US domestic gas production. This partially reflects the fact that Canadian natural gas reserves in the west have higher sulfur contents than typical US gas reserves, hence the higher flaring CO2 rate. And this US EIA dataset does not attempt to account for fugitive losses other than from flaring.

Beyond  the limited information you can find at the UNFCCC National Reports website, no one’s estimates are very good for developing nations.

But in the UNFCCC CRF reports, the estimates for electricity consumption (in Tjs) covers all electricity consumption in some national reports, and only fossil-and biomass-fuelled electricity consumption in others (no nuclear or hydro). So while this is the best place to get good electricity sector GHG estimates, you can’t rely on the CRF reports for total electricity demand or production part of the equation.

The best estimates of total electricity output, consumption, imports and exports, by fuel of origin and including renewables, hydro and nuclear, are here, under "Electricity" and "Renewables". Use the gross heat content tables in these sections to convert GWh and Quadrillion BTUs to Tj or whatever common energy output reporting format you wish to use.

The production, consumption and trade estimates at this US EIA website are simply reprinted International Energy Agency (IEA) data and easier to access and use than the IEA data. But the GHG data at the US EIA site is unreliable (CO2 only for some nations, more or all GHGs for others). The easiest way to put pretty good GHG/MWh-equivalent numbers is to put the GHG estimates from the UNFCCC CRF reports together with the electricity production, consumption and/or trade data from the US EIA website.

You may or may not want to see what it would mean to adjust national GHG liabilities to reflect electricity trade, as the US proposed climate change legislation dictates. Note that the US climate change bills oblige states to book GHGs arising from electricity demand in the state of final consumption, not the state in which generation occurs. This US-dictated GHG reporting procedure potentially delivers huge trade advantages to the US at the expense of Canadian clean electricity exports. We have not yet properly discussed this issue in Canada, at least as far as I know.

The US EIA version of the electricity trade data does not tell you electricity country of origin or destination for exports. If you need to get more into the implications of booking power generation attributes to the importing state/national GHG inventory, and want to assign net GHGs arising from electricity trade to the importing nation, you might want to buy some data directly from the IEA, here. The free data (last year reported 2008) does not tell you enough.. But the IEA emissions reports cover only CO2 and not all GHGs, so while you can use the IEA electricity production trade numbers with confidence, you still want to use the UNFCCC CRF GHG estimates where they are available. 

I find the IEA data too expensive. So for Europe, I use Eurostat production, import and export data (which is consistent with IEA and OECD data), which you can find here, go to "Energy Statistics – quantities (nrg_quant)"  under "Database". Here you can find electricity imports and exports by country of origin and destination for European nations’ electricity consumption.  Again, use the UNFCC GHG data to establish national sector GHG baselines, do not the mixed up CO2/GHG data in the Eurostat database for GHG/MWh baseline national estimates..

For Canada/US electricity trade, you can rely on the National Energy Board electricity trade reports, here. Here you can see trade between provinces and states.  If you want o figure out what GHG factor to assign to trade broken down by province and state:

• for provincial electricity sector GHG factors go to Environment Canada’s national GHG inventory, go here for 2006 and earlier numbers, email request for most recent—2007—report and go down the Table of Contents to click on ANNEX 9: Electricity Intensity Tables).

• for or US electricity sector GHG factors  by state, go here, and do an "all programs" state level emission search under the "unit level emissions" option.

There are still some significant problems with the Environment Canada estimates, but these are still the best official estimates you will get for now. The US numbers are much more reliable.  This fact alone is a significant source of tariff risk for Canada/US electricity trade.

If you do not care about imports and exports by country/state of origin, destination, another good source with a different production breakdown is the set of production import and export estimates from the free IEA tables here(go here, click on excel or archives after "electricity" under "Related Surveys". )

For the most up-to-date comparable data on national energy and environmental taxes and policies, including current energy and environmental tax rates and total government revenues, go to the OECD Economic Instruments database, here.

But remember that it is rarely the case that tax increases are fully passed through as retail price increases on each fuel that government might target with taxes. For example, fuel suppliers might respond to an increase in taxes on gasoline with a reduction in the wholesale price of gasoline and a commensurate increase in the wholesale price of (tax exempt) propane or ethanol. To see clearly how rarely point of production or consumption taxes/tax increases translate into equivalent retail (tax included) prices/price increases, go back to the Eurostat website, here, click on "database", then "environment and energy", then "energy", then "energy statistics – prices" then compare prices with and without all taxes, with and without VAT. And compare rates for residential customers to industrial customers.

Note that energy vendors will shift cap and trade compliance cost burdens in the same way they have always manipulated wholesale fuel prices to move around carbon tax burdens. Ultimately, integrated energy companies will lay off the lion’s share of any new Canadian tax or cap and trade regulation compliance costs on domestic "captive" customers (where there is no risk of customer flight), who will pay a disproportionate share of either cost burden (relative to "footloose" industrial customers and export markets.,  This is the primary reason I opposed carbon/CO2 taxes as a climate change mitigation measure and favour product standards (with credit trading and banking over quota-based cap and trade.