From the report, the $30 billion in 2004 was spread reasonably equally among the alternatives:
An additional $20-25 billion/year goes to large hydropower projects.
The sources responsible for these investments range from government projects to venture capital. The fastest growing sector recently has been solar photovoltaics, in particular grid-connected PV, with 60% compounded annual growth rates from 2000 to 2004. Venture capital investments in solar projects have been doubling every year from 2001 to 2004.
While still not contributing a huge fraction of world energy supply, this level of investment in renewables is already substantial in comparison to traditional power plant capital investments, worldwide estimated at roughly $150 billion/year. Total world energy infrastructure investment (mining, drilling, refining, etc for fossil fuels, mostly) is likely over $500 billion/year, so the new renewable and large hydro investments represent on the order of 10 percent of total energy investment.
More importantly, as Joel Makower's "hockey stick" comment attests, the extremely rapid growth in these areas suggests we may not have to wait as long as we once thought, before they actually are contributing substantially to world energy supply. Part of that is that it's become clear that, under at least some conditions (helped considerably by government subsidies and perhaps also by high oil prices), renewable energy is actually an economically rational investment. Viability without subsidy would trigger even faster growth rates, at least until we're much closer to saturating world energy demands from renewables.