Historic bubbles · Market concentration
Historic Bubbles Market Concentration
Loading validated concentration history…
Historical curves: approximate digitisation of the supplied BofA Global Research exhibit, credited to BofA Global Investment Strategy, GFD Finaeon and Bloomberg. Colours were traced from the image; explicitly printed peaks (63%, 36%, 40%, 44%, 41%) anchor points obscured by the peak circles. Dates are approximate, calibrated separately for each compressed episode. These are not raw vendor observations. In the exhibit, U.S. bubbles are shares of U.S. market capitalisation; Japan is a share of MSCI ACWI. Exact historical basket definitions are unavailable.
AI Big 10: Apple, Microsoft, Nvidia, Amazon, Alphabet, Meta, Tesla, Broadcom, AMD and Micron, relative to float-adjusted S&P 500 equity capitalisation. History uses iShares IVV dated holdings; ongoing updates use State Street SPY daily holdings. Both funds track the same index. Combined basket equity holding values (IVV) or weights (SPY) are divided by all equity values or weights, excluding cash and derivatives. Alphabet includes both share classes; FB and META identify the same company.
Early Tesla history: Tesla joined the S&P 500 on 21 December 2020. To keep the same ten-company numerator throughout, earlier Tesla contributions are estimated from actual, same-date ITOT holdings and the IVV/ITOT denominator ratio implied by the other nine basket companies. These observations are identified in the downloaded data; current shares are never carried backwards.
Comparison: The AI denominator is the S&P 500, a narrower universe than the entire U.S. stock market. This sourced series is not an exact reconstruction of BofA’s undisclosed Bloomberg calculation and is not rescaled to its printed 42%. Same-date IVV and SPY basket concentrations were cross-checked before switching sources; small fund tracking differences can remain. The ten-company basket is fixed retrospectively, not the ten largest stocks at each date. Japan uses MSCI ACWI; historical basket definitions are unavailable.
Updates: the server checks State Street three times daily (06:50, 12:50 and 18:50 New York time). Month-end history begins in January 2018; new dated daily observations are appended automatically. Dates, all ten members, coverage and weights are validated. Failed refreshes retain the last valid data and show a notice. Lines connect actual observations. Historical curves remain fixed.
AI Big 10: Apple, Microsoft, Nvidia, Amazon, Alphabet, Meta, Tesla, Broadcom, AMD and Micron, relative to float-adjusted S&P 500 equity capitalisation. History uses iShares IVV dated holdings; ongoing updates use State Street SPY daily holdings. Both funds track the same index. Combined basket equity holding values (IVV) or weights (SPY) are divided by all equity values or weights, excluding cash and derivatives. Alphabet includes both share classes; FB and META identify the same company.
Early Tesla history: Tesla joined the S&P 500 on 21 December 2020. To keep the same ten-company numerator throughout, earlier Tesla contributions are estimated from actual, same-date ITOT holdings and the IVV/ITOT denominator ratio implied by the other nine basket companies. These observations are identified in the downloaded data; current shares are never carried backwards.
Comparison: The AI denominator is the S&P 500, a narrower universe than the entire U.S. stock market. This sourced series is not an exact reconstruction of BofA’s undisclosed Bloomberg calculation and is not rescaled to its printed 42%. Same-date IVV and SPY basket concentrations were cross-checked before switching sources; small fund tracking differences can remain. The ten-company basket is fixed retrospectively, not the ten largest stocks at each date. Japan uses MSCI ACWI; historical basket definitions are unavailable.
Updates: the server checks State Street three times daily (06:50, 12:50 and 18:50 New York time). Month-end history begins in January 2018; new dated daily observations are appended automatically. Dates, all ten members, coverage and weights are validated. Failed refreshes retain the last valid data and show a notice. Lines connect actual observations. Historical curves remain fixed.