You probably remember the late 1970's better than I do - that was when the Department of Energy was started by Pres. Carter, and their study (with NASA) on Solar Power Satellites makes an excellent reference point on the subject. It was not that which "swatted it down" (although it did make it clear it was going to be a very expensive project with then-current or near-term technology) but a follow-on report from the congressional Office of Technology Assessment, which you can read here:

http://www.wws.princeton.edu/cgi-bin/byteserv.prl/~ota/disk3/1981/8124/812401.PDF

which came out in August 1981. You'll note the president was no longer Carter, which may or may not have had anything to do with the fact that NASA and DOE did nothing on the subject for the next 15 years or so. Reagan had his own way of dealing with the oil crisis, which proved remarkably succesful in the short term.

Anyway, if you read the OTA report, you'll note that even that doesn't "swat it down" so much as say the technology is probably 20 years or so away - by which time fusion power will be commercially available!

The 1996 re-assessment and program re-initiated by NASA was reviewed by the National Academies in 2001, you can read that report here:

http://www.nap.edu/execsumm/0309075971.html

While having some criticisms for NASA, this report was markedly more positive than the 1981 OTA report, and lends a lot of credibility to the concept.

The paranoid will note that NASA's SPS program was again cancelled by a Republican president in 2001.

The fundamental reasons why SPS MIGHT make real sense are two-fold:
(1) If solar photovoltaics are necessarily high-cost materials, placing them in space generates several times as much energy for each dollar of material.
(2) Even if solar PV becomes cheap, if the costs of overnight storage and transmission of energy remain high, the continuous availability and ubiquity of a space solution may actually be cost effective.

Both of these require launch costs to be somewhat lower than they are now - but $1000/kg may be enough, depending on specific power levels achievable. It's hard to know anything more without a really serious R&D program in the area though, and those just seem to keep getting killed off just when they look promising...