История уже случилась. Сыграйте её заново.

Broad equity market return for USA in 2008, in USD, derived in two steps from a LOCAL-CURRENCY index level (§8, §85). (a) local return = idx(Dec 2008)/idx(Dec 2007) - 1 = 51.73662/91.88258 - 1 = -0.436926782. (b) USD return = (1 + -0.436926782) * FX(2007)/FX(2008) - 1 = (1 + -0.436926782) * 1/1 - 1 = -0.436926782, where FX is LCU per USD so a rising FX means the currency weakened. Index source: OECD Short-Term Economic Statistics, Financial market, share price index (MEASURE=SHARE, UNIT_MEASURE=IX), MONTHLY series read at DECEMBER, index level in LOCAL currency. The FX step is applied EXACTLY ONCE. This differs from the v1 series (World Bank CM.MKT.INDX.ZG), which was already a USD percentage change and to which §8 forbids applying FX at all; v1 and v2 therefore disagree by roughly the currency move in any year the currency moved. THE INDEX IS MEASURED IN DECEMBER AT BOTH ENDS: a December-to-December return, which is what the game models (a position held across the year). The monthly figures both the OECD and the IMF publish are averages WITHIN December rather than 31-December closes, so this is not a closing-price return. It is NOT the same convention as `fx_rates.json` and the savings mechanic, which use the ANNUAL-AVERAGE PA.NUS.FCRF rate §8 prescribes; the two numbers will not match, on purpose.

Banking & Finance 2008: factor derived from Kenneth French 49 Industry Portfolios, value-weighted monthly returns for column 'Banks', compounded to annual (column 'Banks' on sheet 'Average Value Weighted Returns -- Monthly'). Excess return is measured against the Kenneth French US market total return (Mkt-RF + RF), compounded to annual, so that every sector is comparable (§11). HONEST LIMITATION (§12B, §98): this is a GLOBAL/US listed-sector cycle, not the return of this industry in the player's country. Confidence is LOW for that reason.

Broad equity market return for USA in 2008, in USD, derived in two steps from a LOCAL-CURRENCY index level (§8, §85). (a) local return = idx(Dec 2008)/idx(Dec 2007) - 1 = 51.73662/91.88258 - 1 = -0.436926782. (b) USD return = (1 + -0.436926782) * FX(2007)/FX(2008) - 1 = (1 + -0.436926782) * 1/1 - 1 = -0.436926782, where FX is LCU per USD so a rising FX means the currency weakened. Index source: OECD Short-Term Economic Statistics, Financial market, share price index (MEASURE=SHARE, UNIT_MEASURE=IX), MONTHLY series read at DECEMBER, index level in LOCAL currency. The FX step is applied EXACTLY ONCE. This differs from the v1 series (World Bank CM.MKT.INDX.ZG), which was already a USD percentage change and to which §8 forbids applying FX at all; v1 and v2 therefore disagree by roughly the currency move in any year the currency moved. THE INDEX IS MEASURED IN DECEMBER AT BOTH ENDS: a December-to-December return, which is what the game models (a position held across the year). The monthly figures both the OECD and the IMF publish are averages WITHIN December rather than 31-December closes, so this is not a closing-price return. It is NOT the same convention as `fx_rates.json` and the savings mechanic, which use the ANNUAL-AVERAGE PA.NUS.FCRF rate §8 prescribes; the two numbers will not match, on purpose.

Technology 2008: factor derived from Kenneth French 49 Industry Portfolios, value-weighted monthly returns for column 'Hardw', compounded to annual (column 'Hardw' on sheet 'Average Value Weighted Returns -- Monthly'). Excess return is measured against the Kenneth French US market total return (Mkt-RF + RF), compounded to annual, so that every sector is comparable (§11). HONEST LIMITATION (§12B, §98): this is a GLOBAL/US listed-sector cycle, not the return of this industry in the player's country. Confidence is LOW for that reason.

Deposit rate for USA 2008 substituted by the country's OWN short-term money-market rate from OECD Short-Term Economic Statistics (MEASURE=IR3TIB), because neither the World Bank (FR.INR.DPST) nor the IMF publishes a deposit rate for this country in any year of the period. AN INTERBANK RATE IS NOT A DEPOSIT RATE: it is the rate banks lend to each other at, and it normally sits above what a retail saver is paid, so this OVERSTATES the return on a deposit. It is used unadjusted — applying a haircut would be inventing a coefficient. It is preferred over the regional median regardless, because it is an observation OF THIS COUNTRY: the median gave Denmark, the United Kingdom and Iceland identical rates for 26 straight years, which erases the currency choice the savings mechanic is built on (§80, §81).

USD является базовой учётной валютой игры, поэтому её курс к самой себе тождественно равен 1, а доходность — 0.

Interest is ordinary income in the US. Top federal marginal rate on ordinary income. Federal only: state and city income taxes are not modelled, so an actual New York or California resident faced a materially higher combined rate.

Household final consumption expenditure per capita (current US$), USA 2008, computed as NE.CON.PRVT.CD / SP.POP.TOTL. Living-cost proxy per §9, CONSTRUCTED rather than read from a single indicator: household and NPISH final consumption expenditure in current US$ (World Bank WDI NE.CON.PRVT.CD) divided by total population (World Bank WDI SP.POP.TOTL), the division carried out in exact decimal (§35). The result is what households in that country actually spent per person in the dollars OF THAT YEAR, so it carries the country's own inflation and its own exchange rate, and is therefore in the same units as the player's capital, the market returns, the deposit rates and the FX in this dataset — all of which are nominal. The World Bank's own per-capita series NE.CON.PRVT.PC.KD is in CONSTANT 2015 US$ and is deliberately NOT used: it charged 2015 purchasing power against nominal money, over-stating the cost of living by about 38% in 2000 and under-stating it by about 26% in 2025. No per-capita current-US$ indicator is published (NE.CON.PRVT.PC.CD does not exist), hence the division. A country-year exists only when BOTH series report it. The figure is an economy-wide average of what households actually spent per person, not a cost-of-living survey; the engine scales it by household size and a documented gameplay factor (§9).