What's the difference between EBITDA and AUM as a valuation metric? How can the prices paid for fund sponsors in the late 1990's be justified in terms of either metric?

Valuing fund sponsors under AUM (assets under management) is mathematically simple, but often misleading. The mathematics involve dividing the total purchase price by the total assets managed by the acquired firm to product a percentage – e.g., this firm was bought for 3% of assets under management. But AUM can be misleading because asset managers have different revenue generation and profit margins depending on customer and product segments – for example, institutional money market funds versus retail equity funds.
EBITDA means earnings before interest, taxes, depreciation and amortization; it is a pretty good measure of cash flow after operating expenses. Thus, many acquisitions of fund sponsors are priced as a multiple of EBITDA, much like stocks are valued on price/earnings ratios. These acquisitions have been priced from single-digit multiples of EBITDA to multiples in the 20's.
In order to justify the high premiums paid for fund sponsors, whether measured as a percentage of AUM or as a multiple of EBITDA, acquirers must assume that the high growth rate of these sponsors in the 1990s can be extrapolated into the next decade. The profits from the sponsor's past and currently expected earnings are already built into the price. However, it remains to be seen whether such high growth rates can be maintained in the fund industry, especially in light of the recent struggles of the U.S. equity markets. Moreover, it is unclear whether the acquirers can meet the challenge of retaining and motivating investment talent in the acquirer organization.