1. dogma1138, your own cite doesn't support your "insurance and pension is > 50%" figure. The Bloomberg link shows a 2014 graphic where: Public pension: 20

1. dogma1138, your own cite doesn't support your "insurance and pension is > 50%" figure. The Bloomberg link shows a 2014 graphic where: Public pension: 20% Corporate pension: 7% Insurance companies: 7% Union pension funds: 2% --- "Insurance and pension" total: 36% (This is the fraction of the VC's fund that is committed by those types of institutions, NOT how much of those institutions' portfolios are in VC; see below for that.) 1.5. Elsewhere in this thread, dogma1138 suggests that the Bloomberg mention of CALPERS trimming their VC allocation is "to 1 percent, from 7 percent..." What is left off is the crucial kicker: "...of its private equity portfolio." Meaning that even CALPERS is allocating only 0.5% or so to ALL its venture capital managers. 2. The fraction of an institutional portfolio that goes to any one VC fund is usually well under 1%. How it actually works (foundation / endowment model): - Estimate asset class returns, volatilities, and covariances. This is done with historical analysis and some forward-looking hand-wave mumbo-jumbo. - Chart out the "efficient frontier" of expected return per unit "risk" (volatility of return) across several model portfolios (each of which is a set of % allocations to the various asset classes). This makes a nice hyperbola that looks like a "C" and you pick one of the ones in the top left, meaning, more return and less risk. - That asset allocation will almost certainly look something like: Public Equity: most Bonds: second-most Hedge Funds: small Private Equity: small "Alternatives" in vogue at the time (timber, gold, real estate, rice paddies, comic book royalties, whatever): small (Where "small" usually means < 15%) - Then, within the "Private Equity" bucket, you divvy that up and do a similar exercise: Large Buyout: most Middle-Market: recently more fashionable Venture Capital: small (Here, the VC part of private equity is likely 0-25%, unlikely to be a big part of the PE bucket.) - Then, within each of those buckets, you pick a variety of "managers." Probably at least 3-5 in the VC bucket. - Then, with each VC "manager" (VC firm) you allocate your commitments such that you can have some continuity of investment across the serial funds of that firm. (Note: if you don't have the size or sophistication to manage managers, you might just buy a "fund of funds" to achieve diversification within the VC sub-allocation of your PE allocation). If an institution has 10% allocated to PE, they might have 2% allocated to VC, and of that, 0.5% might be allocated to a particular manager (firm), and it might well be put no more than 0.25% into a particular fund. 2.5 Insurers and pensions work differently from the above (endowment/foundation model) in some ways, the same in other ways. Crucially they both have to manage the expected liabilities of near term payouts -- but they don't do that with private equity / venture investments, they do it with bonds. See "immunization." They do, however, try to generate a "total return" on the surplus that is not expected to be used near term. Also I'm leaving out minimum 5% spend, etc., I know. 3. It is totally rational to put a small but meaningful sliver of your large portfolio into venture capital, especially and crucially at a time when entire industries (e.g. taxicab, booksellers, maybe internal combustion cars??) can get decimated by technology disruption within a decade. If nothing else, it's like buying a cheap put against the old-line stocks in your portfolio. 4. Yes, VC has its issues but dogma1138 is not pointing out any of them. So, I'm sorry, but you're completely out to lunch if you think a VC fund or private equity fund is going to cause an insurer or pension "even 5-10% loss [such that] tons of people lose coverage, premiums sky rocket," etc.