Everybody would be paying the extra rent (and less in the utility bills), so all the landlords would start out equal; it's just a question of how much it costs the landlord to deliver the X BTU/month, and thus how much of the rent they get to keep.

I expect wide regional variations in such laws; what matters is that they apply to all equivalent units in a market.

"You're going to require landlords to spend a bunch of money and raise rents? What about the people who complain about the lack of affordable housing?"

Does it really raise rates, if the greater efficiency pays for itself? Don't most of these pay for themselves, if you amortize the cost over the life of the improvement?

Much higher energy prices makes it all happen automatically. It's just a painful mechanism.

Only if the customers are well informed and rational. The penetration of CFL here in Northern California, where many are paying $.35 KWhr (marginal rates for people who use a lot), is still pretty low (under 50%). Just having cost effective solutions available, doesn't mean they will be widely utilized.

Then think of the typical industrial situation. Costs are compartmentalized. One department is responsible for capital expenditures, another for operating costs. If the manager of the former buys a premium priced unit which will pay for itself by cutting the operational budget, he may still have to answer to the COO for exceeding his capital budget! There is a lot of inertia, both institutional, and psychological. This means that the actually deployed systems are far behind current (economics tuned) optimal solutions.

You're going to require landlords to spend a bunch of money and raise rents? What about the people who complain about the lack of affordable housing?

They're already paying for the energy costs (or doing completely bone-headed things like using and refueling kerosene heaters indoors, with occasionally fatal results).  This would be a cost shift, not a cost increase.