Methodology & Data: CDs vs. Gold, Silver and Goldbacks
Methodology & Data: CDs vs. Gold, Silver and Goldbacks
This page contains the supporting data, assumptions and calculation methods used in our article:
CDs vs. Precious Metals: A 20-Year Study—and the New Club Partner Program It Helped Inspire
The main article was intentionally written to be readable rather than burying readers in spreadsheets. This page is for anyone who wants to look under the hood, check the assumptions and see how the comparisons were calculated.
Lease-scenario update: October 2, 2026. The original 2% baseline is retained. Additional 4% and 6% scenarios illustrate the effect of different lease rates on selected historical examples.
The Basic Rule: Same Money, Same Starting Line, Same Finish Line
Each example begins with a hypothetical $10,000.
For a five-year comparison, the same $10,000 is assumed to go into either:
- a five-year bank CD at the national-average bank rate available near the beginning of that period, or
- gold,
- silver, or
- a modeled Goldback position.
The investment is then measured over the same five-year period.
This avoids comparing yesterday’s gold market with today’s CD rate.
How the CD Examples Were Calculated
For the CD comparisons shown in the main article, we used published national-average five-year bank CD rates near the beginning of the corresponding period.
The formula is:
Ending CD value = $10,000 × (1 + APY)5
The examples used in the article are:
| Five-Year Period | Bank 5-Year CD Rate at Start | $10,000 After 5 Years |
|---|---|---|
| 2009–2013 | 3.12% | $11,660 |
| 2011–2015 | 1.80% | $10,933 |
| 2015–2019 | 1.18% | $10,604 |
| 2020–2024 | 1.59% | $10,821 |
The historical CD rates above come from National Credit Union Administration data comparing national-average bank and credit-union deposit rates.
These figures assume the CD is held to maturity and that the stated annual yield remains applicable for the entire five-year term.
Individual banks and credit unions could have offered rates above or below these national averages.
How Gold and Silver Were Calculated
Gold and silver use published year-end market prices.
For each five-year period:
Metal return = ending metal price ÷ beginning metal price − 1
For example, the 2020–2024 period begins with the year-end 2019 price and ends with the year-end 2024 price.
Gold:
$2,623.84 ÷ $1,517.39 − 1 = approximately 72.9%
Silver:
$28.90 ÷ $17.83 − 1 = approximately 62.1%
Historical year-end prices used in this analysis came from published gold and silver market-price histories.
How the Lease Was Modeled
In an October 2, 2026, communication to K2BX, Dennis Keating of Alpine Gold Exchange indicated that current lease rates range from 2% to 6%.
This update replaces the previous lease-value tier table on this page. The reported range does not, by itself, establish the balance requirements, eligible metals, terms or other qualifications for each rate.
Readers should confirm the rate and conditions applicable to their proposed lease directly with Alpine Gold Exchange.
For the historical model, we use three hypothetical annual rates:
| Assumed APR | Role in This Study |
|---|---|
| 2% | Lower end of the reported range; retained as the original baseline. |
| 4% | Illustrative arithmetic midpoint; not a verified offer for a particular account. |
| 6% | Upper end of the reported range; an additional hypothetical scenario. |
The $10,000 starting amount is a common measuring scale. It does not mean a $10,000 lease qualifies for all three rates, or that every metal is eligible at each rate.
The model treats each scenario rate as a nominal annual percentage rate, or APR, and assumes:
- The chosen rate remains constant throughout the five-year period.
- Lease returns accrue monthly at one-twelfth of the annual rate.
- All monthly returns are immediately reinvested in the modeled metal position.
- Reinvested returns earn the same assumed rate.
- There are no interruptions, additional contributions or withdrawals.
These assumptions extend the monthly-reinvestment approach used in the original study. They are not a statement that Alpine guarantees uninterrupted reinvestment or any particular rate for five years.
For current program information, see Alpine Gold Exchange’s lease page and confirm the actual agreement terms directly with Alpine.
Five-Year Lease Multipliers
For an annual rate r, expressed as a decimal:
Lease multiplier = (1 + r ÷ 12)60
| Assumed APR | Five-Year Multiplier | Modeled Quantity Increase | $10,000 if Metal Price Is Unchanged |
|---|---|---|---|
| 2% | 1.10507893 | 10.51% | $11,051 |
| 4% | 1.22099659 | 22.10% | $12,210 |
| 6% | 1.34885015 | 34.89% | $13,489 |
The multipliers describe the assumed accumulation from reinvested lease returns. They do not include changes in the metal’s market price.
The complete historical leased-metal formula is:
Ending value = $10,000 × (ending metal price ÷ beginning metal price) × (1 + r ÷ 12)60
Equivalently:
Ending value = $10,000 × (1 + metal return) × lease multiplier
The price factor is the ending price divided by the beginning price—not the percentage gain alone. For example, a 50% price gain uses a factor of 1.50.
This is a hypothetical back-test. Alpine’s current lease structure did not exist throughout the entire historical period, and the results are not actual historical Alpine account statements.
All 16 Rolling Five-Year Precious-Metal Periods
The main study uses every complete five-year period beginning with 2005–2009 and ending with 2020–2024.
The following table retains the original 2% APR baseline. The expanded 4% and 6% comparisons follow it.
| Period | Gold Price Change | Silver Price Change | $10,000 Gold + 2% Lease Model | $10,000 Silver + 2% Lease Model |
|---|---|---|---|---|
| 2005–2009 | +150.0% | +147.1% | $27,629 | $27,302 |
| 2006–2010 | +174.6% | +250.5% | $30,348 | $38,728 |
| 2007–2011 | +145.7% | +116.2% | $27,146 | $23,896 |
| 2008–2012 | +100.9% | +106.0% | $22,204 | $22,762 |
| 2009–2013 | +37.0% | +72.4% | $15,136 | $19,048 |
| 2010–2014 | +8.0% | −6.5% | $11,935 | $10,334 |
| 2011–2015 | −25.3% | −55.2% | $8,251 | $4,952 |
| 2012–2016 | −26.4% | −42.5% | $8,137 | $6,359 |
| 2013–2017 | −22.2% | −44.1% | $8,594 | $6,174 |
| 2014–2018 | +6.4% | −20.3% | $11,759 | $8,802 |
| 2015–2019 | +28.2% | +13.5% | $14,163 | $12,542 |
| 2016–2020 | +78.7% | +90.3% | $19,747 | $21,032 |
| 2017–2021 | +58.7% | +46.2% | $17,541 | $16,160 |
| 2018–2022 | +40.0% | +41.4% | $15,471 | $15,621 |
| 2019–2023 | +60.8% | +53.5% | $17,771 | $16,968 |
| 2020–2024 | +72.9% | +62.1% | $19,109 | $17,912 |
The original study used annual gold and silver closing-price histories to support these rolling calculations. One referenced source was StatMuse’s annual gold-price history.
Expanded Comparison: 2%, 4%, and 6%
To keep the main article readable, the expanded comparison uses three examples for each metal:
- The weakest five-year price-performance period in the main study.
- The strongest five-year price-performance period in the main study.
- The most recent five-year period included in the main study, 2020–2024.
These examples are selected by those rules, rather than by choosing only favorable outcomes.
Gold: Approximate Five-Year Ending Values
| Historical Example | 2% APR | 4% APR | 6% APR |
|---|---|---|---|
| Weakest: 2012–2016 | $8,137 | $8,991 | $9,932 |
| Strongest: 2006–2010 | $30,348 | $33,531 | $37,043 |
| Most recent: 2020–2024 | $19,109 | $21,113 | $23,324 |
Silver: Approximate Five-Year Ending Values
| Historical Example | 2% APR | 4% APR | 6% APR |
|---|---|---|---|
| Weakest: 2011–2015 | $4,952 | $5,471 | $6,044 |
| Strongest: 2006–2010 | $38,728 | $42,790 | $47,271 |
| Most recent: 2020–2024 | $17,912 | $19,791 | $21,863 |
All figures begin with $10,000 and include both historical metal-price changes and hypothetical reinvested lease returns. They are ending values, not lease income alone.
Even the 6% scenario does not prevent a loss in the weakest gold and silver examples.
How the Additional Scenarios Were Calculated
The original table reports rounded whole-dollar ending values. For this update, the new scenarios were derived from those retained 2% values rather than reconstructed from a complete set of unrounded historical price inputs.
The adjustment is:
Adjusted ending value at rate r = reported 2% ending value × [(1 + r ÷ 12)60 ÷ (1 + 0.02 ÷ 12)60]
This removes the original 2% lease multiplier and substitutes the selected scenario multiplier while retaining the original modeled metal-price result.
For example, the weakest gold period at 6% is calculated as:
$8,137 × [(1 + 0.06 ÷ 12)60 ÷ (1 + 0.02 ÷ 12)60] ≈ $9,932
Full-precision multipliers are used for the calculation, and final results are rounded to the nearest dollar. Because the starting 2% values were already rounded, the new figures are approximate and may differ slightly from a fresh calculation using the original unrounded price inputs.
This update expands the lease-rate scenarios; it does not replace or independently re-audit the original historical dataset.
Summary of the 16 Historical Periods
Before lease income:
| Five-Year Result | Gold | Silver |
|---|---|---|
| Highest | +174.6% | +250.5% |
| Average | +55.5% | +51.9% |
| Median | +49.4% | +49.9% |
| Lowest | −26.4% | −55.2% |
The median is particularly useful because it is less affected by unusually strong or unusually weak periods.
The average can be pulled upward substantially by years such as 2006–2010.
The worst periods are included deliberately because the purpose of the study is not to show only favorable precious-metal results.
Changing the hypothetical lease rate does not change these underlying metal-price statistics.
How Goldbacks Were Modeled
Goldbacks did not exist throughout this historical period.
Goldback began in 2019, so there is no actual 20-year Goldback market-price history.
For purposes of this study only, we use:
Modeled 1 Goldback value = gold spot price × 0.002
A 1 Goldback denomination contains 1/1,000 of a troy ounce of gold.
Our model therefore assumes a value equal to twice the spot value of its contained gold.
Because that multiple remains constant throughout the model, the modeled Goldback has exactly the same percentage price change as gold.
For example:
If gold increases 50%, the modeled Goldback increases 50%.
If gold falls 20%, the modeled Goldback falls 20%.
The model does not attempt to reconstruct actual historical Goldback premiums, manufacturing costs, Daily Exchange Rates, secondary-market pricing or collector premiums.
Pre-2019 Goldback figures should therefore be understood strictly as a mathematical back-cast.
The expanded lease tables are labeled for gold and silver. They do not establish Goldback lease eligibility or imply that each scenario rate is available for Goldbacks.
Why 2025 and 2026 Were Excluded
The primary study deliberately stops at December 31, 2024.
This was not because 2025 was unfavorable to precious metals.
It was exactly the opposite.
Gold rose dramatically in 2025, and silver rose even more dramatically.
Including the completed 2021–2025 period adds the following five-year result:
| 2021–2025 | Price Change | $10,000 + 2% Lease Model |
|---|---|---|
| Gold | +127.5% | about $25,144 |
| Silver | +167.3% | about $29,539 |
This supplemental table retains the original 2% baseline and is not included in the 16-period main study.
Adding that single strong period changes the historical averages materially.
| Statistic | Through 2024 | Including 2025 |
|---|---|---|
| Gold average 5-year gain | 55.5% | 59.7% |
| Gold median 5-year gain | 49.4% | 58.7% |
| Silver average 5-year gain | 51.9% | 58.7% |
| Silver median 5-year gain | 49.9% | 53.5% |
That is why it was excluded from the primary study.
We wanted the main article to answer the question without allowing an unusually strong recent precious-metals market to dominate the result.
Year-to-date 2026 is also excluded because it is not a completed calendar year.
What the Model Does Not Include
To keep the comparison understandable, several variables are intentionally excluded.
The model does not attempt to include:
Taxes.
Tax treatment varies depending on the organization and its legal and tax classification.
Early CD withdrawals.
The CD examples assume the CD is held to maturity.
Buying above or below national-average CD rates.
An organization shopping aggressively may find a better rate than the national average.
Metal transaction spreads or premiums.
Actual acquisition or liquidation costs can affect returns.
Changes in lease rates.
Each scenario applies a constant hypothetical 2%, 4% or 6% APR for five years. Actual rates and renewal terms may change.
Changes in lease eligibility or capacity.
Programs and availability can change over time. The model does not determine whether a particular account or metal qualifies for a scenario rate.
Reinvestment interruptions, program fees or payment shortfalls.
The model assumes all lease returns are received and immediately reinvested at the same rate without deductions or delays.
Inflation.
The primary figures are nominal dollars rather than inflation-adjusted purchasing power.
Goldback collector premiums.
The modeled Goldback follows gold mathematically and does not attempt to value collectible or scarcity premiums.
Why We Used $10,000
There is nothing special about $10,000.
It was selected because it makes the math easy to understand.
A reader can roughly double the figures for $20,000, multiply them by five for $50,000, and so forth, provided the same modeling assumptions are used.
That mathematical scaling is separate from actual program eligibility. A different lease amount may qualify for different terms, and readers should confirm applicable rates and requirements directly with Alpine.
Historical CD Sources
The article's CD examples use national-average bank rates rather than today's rate applied backward through history.
For example:
- December 2008 national-average five-year bank CD: 3.12%
- December 2010: 1.80%
- December 2014: 1.18%
- December 2019: 1.59%
NCUA's published historical rate tables document those national averages. See the NCUA fourth-quarter 2008 table as an example.
For broader historical context, the Federal Reserve Bank of St. Louis also publishes Bankrate's long-running five-year CD series, BRMCDS0102, which tracks national-average five-year CD APYs and is intended to provide consistent rate comparisons over time. See FRED.
Sources Used
National Credit Union Administration (NCUA)
Historical Credit Union and Bank Rates tables.
Federal Reserve Bank of St. Louis / FRED
Bankrate Monitor five-year CD APY series, BRMCDS0102.
Published Gold and Silver Market-Price Histories
Annual year-end gold and silver closing-price data used in the original study.
Alpine Gold Exchange
Precious Metal Leases and Rates & Fees information referenced in the original study.
October 2, 2026, Communication from Dennis Keating, Alpine Gold Exchange
Source for the reported current 2%–6% lease-rate range. The 4% scenario is an illustrative midpoint selected for this comparison.
A Final Note About Back-Testing
Back-testing can answer:
“What would this model have done during these historical periods?”
It cannot answer:
“What will happen during the next five years?”
Gold, silver, CD rates, lease rates, inflation and market conditions will all change.
The purpose of this methodology page is not to create a prediction.
It is simply to make the assumptions behind the article transparent enough that readers can see exactly what was—and was not—included in the comparison.