Gold as an Inflation Hedge: Whose Inflation, Which Index

Gold protects you from inflation. The sentence appears on bullion dealer guides, on bank research pages and in newspaper columns, and it is repeated often enough to feel settled. It is not settled, and the reason has little to do with gold.

The sentence leaves out everything a reader needs to check it. It does not say whose inflation, measured by which index, against which gold price, over what stretch of time. Fill those in differently and the same claim comes out true or false. What follows is what each blank does to the answer.

What the Hedging Claim Actually Says

Stated carefully, the claim is an arithmetic one. Over some period, the percentage change in the price of gold was at least as large as the percentage change in a general price level, so a holder ended the period able to buy the same basket as before. Anything short of that is a partial hedge, and anything above it is a real gain that the word hedge does not describe.

Written that way, four inputs appear that the everyday version hides. A currency, because a price level belongs to an economy and a price belongs to a unit of account. An index, because general price level is a construction and different constructions give different answers. A gold price, because more than one is published. And a period, because a claim with no start and end date cannot be wrong.

Nothing here is a criticism of gold. It is a criticism of a sentence. The same four blanks would have to be filled to test the claim for property, equities or anything else, and until they are filled the discussion is about wording rather than about markets. Our page on what the CPI report measures covers how one of those price levels is put together and released.

Whose Inflation: Gold Prices Form in One Currency

The London benchmark is administered by ICE Benchmark Administration, which operates electronic auctions for spot unallocated gold held in London. The administrator states that price formation in those auctions happens in US dollars, and that although the resulting benchmark is also published in pounds and euros, those currency versions are not tradeable directly through the auction itself.

That single fact reorganises the question. A reader whose salary, rent and shopping are denominated in dollars can ask a clean question: did the dollar price of gold keep pace with dollar prices generally. A reader in the euro area, the United Kingdom or the Gulf cannot ask that question, because the number they care about is not the one the auction produces.

For them the gold return has two moving parts. One is what happened to gold against the dollar. The other is what happened to their own currency against the dollar over the same period. Those two can pull in the same direction and flatter the result, or pull against each other and cancel out most of it.

The hedging question they are actually asking is therefore whether the combined outcome kept pace with prices in their own economy. That is a different question from the one the dollar reader asks, and it has a different answer.

This is also why a currency peg matters more than it looks. A reader whose currency is pegged to the dollar has removed one of the two moving parts, but has imported the dollar leg of the calculation into an economy whose own inflation is set locally. The gold price behaves as it does for a dollar holder while the price level does not.

How gold and the dollar move against each other is a separate subject with its own conditions, and it is developed on our page about the gold and dollar relationship rather than repeated here. The point for this page is narrower: the currency you spend decides which calculation you are entitled to run.

Which Index: the 2 Percent Target Is Not Measured on CPI

Discussions of gold and inflation almost always reach for the 2 percent figure at some stage, usually as the level things are supposed to return to. That figure has a precise home, and it is worth knowing what sits inside it.

The Federal Reserve Board answers that directly. A 2 percent objective for the longer run was set out by the Federal Open Market Committee in the document titled Statement on Longer-Run Goals and Monetary Policy Strategy.

The committee tied that objective to one particular series rather than to inflation at large: the price index covering personal consumption expenditures, usually shortened to the PCE price index. What the objective tracks is the yearly movement in that one series.

So the 2 percent anchor is defined on one index. The hedging question, when anyone tries to answer it with data, is almost always run against a different one, the consumer price index. The two are built from different baskets and different weighting rules, and they are not interchangeable inputs into the same sentence.

A reader does not need to prefer one index over the other. What they need is to notice when a single paragraph uses one index to set the target and the other to judge whether gold cleared it. That paragraph has changed the yardstick halfway through, and the conclusion it reaches is about the switch as much as about gold.

The distinction between a nominal quantity and the same quantity adjusted for prices runs through more than this topic. It is what separates a headline interest rate from what a holder of that currency actually receives, and our page on reading real yields instead of headline rates works through that case.

Which Gold Price: Two Daily Settings and a Spot Market

The phrase the gold price suggests one published number. The market does not produce one. ICE Benchmark Administration runs the gold auction twice each London trading day, at 10:30 and at 15:00 London time, and publishes the two outcomes as separate benchmark settings, an AM setting and a PM setting.

Alongside those settings a continuous over-the-counter spot price trades through the day, and it is that spot price, not the benchmark, that a retail trading platform quotes. Three different numbers, all correct, all describing gold.

For a long-horizon purchasing power calculation the choice between them changes very little. For a short-horizon comparison it can change the sign of the answer, because a benchmark setting is a point in an auction at a fixed minute while a spot quote is wherever the market happened to be.

A claim that compares an inflation figure with a gold return has picked one of these three, whether or not it says so. The versions that do not say which have left the reader unable to reproduce the number.

Why Two Page-One Answers Contradict Each Other

Search this question and the first page of results carries both answers at once. A large bank publishes a piece saying gold may be the best available hedge against inflation. A newspaper column, published later, says gold has been a poor short-term hedge. Neither is a fringe source and neither is obviously careless.

They are answering different questions. One is arguing about long horizons, where the comparison is between an ounce of metal and a basket of goods across decades. The other is arguing about short ones, where gold responds to a great deal besides the price level and can move against inflation for years at a stretch. Read as answers to the same question they look like a dispute. Read as answers to two questions they are consistent.

The dates matter as much as the horizons. Of the pages ranking for this question, some carry no publication or update date at all, and at least one leans on figures gathered several years before the reader arrives. A conclusion about an asset that responds to interest rates and currencies is a conclusion about a period, and a page that hides its period is asking to be trusted rather than checked.

The useful move is not to pick a winner between them. It is to ask any such piece which horizon it measured and which years it covered, and to treat an answer that supplies neither as an opinion rather than a finding. Our page comparing gold traded as a store of value against an industrial metal shows how differently the same commodity label can behave depending on what is being asked of it.

What This Page Does Not State, and Why

No inflation rate appears above. No gold price appears above. No historical return, no percentage of years in which gold outpaced prices, and no comparison of gold against equities or bonds appears above either. That is deliberate and it is worth explaining, because those figures are exactly what the competing pages lead with.

Every figure of that kind found while researching this page was either stated with no source at all, or traced back to a single privately compiled dataset rather than to the agency that produces the underlying series. A figure repeated across several pages is still one figure, and repetition is not verification.

The primary sources that would settle them were checked directly. The benchmark administrator and the central bank documentation cited here answered. The statistical agency publishing the US consumer price series did not respond to repeated requests during the checks for this page, so nothing that would have depended on it has been written as though it had.

A reader who wants those numbers can obtain them from the agencies named in the sources line below, in the currency and over the period they choose. That is a better outcome than taking a number from a page that cannot say where it came from, and the difference between the two is the whole subject of this section.

Who This Page Is Not For

Anyone who arrived looking for a recommendation on whether to hold gold will not find one here, and none is being withheld. Whether a particular holding suits a particular person depends on circumstances this page has no access to, and answering it would be advice rather than education.

Nor is this a forecast. Nothing above says what gold or any price level will do next, and the argument is deliberately built out of definitions and published documents rather than out of expectations.

Which part of it applies to you depends on the question you brought. If you trade gold against the dollar intraday, the currency section is the one that matters, and the inflation index barely enters the calculation on that timescale. The practical distinction for you is between the benchmark settings and the spot quote you are filled at.

If instead you are asking a multi-year question about what savings will buy later, the index and the period are decisive and the intraday quote is noise. The two readers share a keyword and almost nothing else, and most of the disagreement about gold and inflation comes from them being answered as though they were one person. Our page on price levels and exchange rates is the next step for the second question.

Sources checked 16 August 2026: ICE Benchmark Administration, LBMA Precious Metals, for the auction times, the two daily gold benchmark settings, the US dollar price formation and the non-tradeable pound and euro publications · Federal Reserve Board, Why does the Federal Reserve aim for inflation of 2 percent over the longer run, for the longer-run objective being tied to the PCE price index rather than to consumer prices · London Bullion Market Association, LBMA Precious Metal Prices, for the London market standards under which the benchmark metal is delivered