Fiat Currency, Inflation, and Why Bitcoin's Fixed Supply Matters

One side of this can print more of itself whenever the pressure builds. The other side cannot, by design, no matter who wants it to. Fiat currencies lose purchasing power over time because a supply that can always grow eventually does. Bitcoin caps itself at 21 million coins on a schedule nobody can amend. For South Africans who have watched the rand slide against the dollar for the better part of a lifetime, that difference stops being a monetary theory and starts being personal.

The word "fiat" is Latin for "let it be done." A fiat currency is worth something because a government declares it worth something, not because it is backed by any physical thing you could weigh in your hand.

That was not always how money worked. For most of recorded history money was tied to a commodity, and the gold standard that linked paper notes to fixed weights of gold acted as a brake on how freely a government could spend. The brake came off in stages. The United States made its final break from gold in August 1971, when President Nixon ended the dollar's convertibility and closed the Bretton Woods era. Every major currency since then has been, in principle, printable without limit. Milton Friedman spent a career arguing that inflation is always and everywhere a monetary phenomenon, that when the supply of money grows faster than the supply of goods and services, prices rise. It is not a conspiracy. It is arithmetic, and it is the arithmetic every saver ends up paying for.

What money printing looks like at scale

The past two decades ran that experiment at a scale nobody had attempted before. Around the pandemic the United States grew its M2 money supply by more than 40% in under three years. The European Central Bank, the Bank of England and the Bank of Japan ran their own versions of the same playbook, and emerging market central banks joined in where they could.

Nobody voted on it. The consequences arrived anyway.

When economies locked down in 2020 the fresh money did not immediately produce more goods. It sat in the system waiting. When supply chains loosened over 2021 and 2022 the price increases landed everywhere at once, on petrol, food, building materials, medical aid and insurance, and the rand in every savings account quietly bought less than it had the year before. I trace what that slow leak does to a lifetime of bank savings in what is actually happening to your savings. The short version is that this is the system working as designed, even if it was never advertised that way.

The South African version, in numbers

South Africans do not need this explained in the abstract. We have lived it.

Inflation here has averaged 8.47% a year since 1968 and touched 20.7% in January 1986. The recent decade looked calm by comparison, and still consumer prices rose 65% between 2010 and 2020, an average of 5.2% a year, across a stretch the Reserve Bank counts among its successes. Nobody felt robbed in any single month. Anyone who left long-term savings in cash across those ten years still lost more than a third of what it could buy. As I write in mid-2026 the official rate has climbed back to 4.5%, driven mostly by fuel, with petrol up nearly 25% over the year and diesel far worse. A client near George put it plainly last winter: his medical aid, his rates and his fuel had all outrun his salary for three years running, and his money market account had not come close to keeping up.

Then there is the part almost nobody reads. In March 2020, when the local bond market seized, the SARB announced that it would create money to buy government bonds, and it began buying on 20 March, five days before the formal announcement. By the end of October 2020 it had accumulated R38.8 billion in government paper, and its balance sheet had grown by almost 15% in that single month of March. The Bank has always insisted this was market functioning rather than quantitative easing. On the mechanics I take the point. The lesson survives the label. Under enough stress the constraint moved, and a sound money is precisely the one whose constraint cannot.

The 2024 GFECRA arrangement is the same story in a quieter suit. That account records the revaluation gains on the country's gold and foreign exchange reserves. It sat at minus R28 billion in 2003. Two decades of rand weakness pushed it beyond R500 billion by 2024, and R150 billion of it is now being paid to Treasury across three years, funded not by selling any reserves but by creating new central bank liabilities. Economists who study the plumbing note that the balance sheet outcome mirrors quantitative easing whatever it is called. By October 2025 broad money was growing at 7.5% a year, with R22.4 billion of a single month's expansion coming from new claims on the government sector.

A revaluation gain that large is not a trading profit. It is twenty years of currency slide, recorded neatly by the state's own accountants. I set out that longer slide in what the rand's slide against the dollar means for Bitcoin.

Why Bitcoin's fixed supply is not a minor technical detail

Bitcoin's entire monetary policy fits in one sentence. There will only ever be 21 million coins and the rate at which new ones appear halves roughly every four years until issuance stops.

Satoshi Nakamoto launched the network in January 2009 and wrote that cap into the protocol not as a policy anyone could revise but as a rule enforced by every full node running the software. New coins enter through mining. Miners earned 50 per block in 2009. After the April 2024 halving they earn 3.125, and by the early 2030s it will be under one. More than nineteen of the twenty-one million coins have already been issued, so the world has met most of the supply it will ever get. No central bank can vote to expand it. No crisis loosens it. No committee sits behind it, because there is no committee at all.

That is the whole difference, and for someone raised where the currency has been debased for decades it is not a small one.

People reach for the word "unbacked" here and get it exactly backwards. Bitcoin is not backed by a promise that can be broken under pressure. It is backed by a rule that a promise cannot override. Where a central bank's constraint is the judgement of the people who benefit from loosening it, Bitcoin's constraint is a line of consensus that rejects any block trying to mint more than the schedule allows, automatically, with nobody to lobby and no appeal to file. That is why the halving has never once slipped. A demand spike does not accelerate it. A geopolitical shock does not pause it. The schedule simply runs, block after block, indifferent to what anybody would prefer it to do.

What scarcity actually buys you

Scarcity on its own is worthless. A limited run of anything nobody wants stays cheap. What gives Bitcoin's scarcity teeth is everything it arrives attached to. There will only ever be 21 million. Appetite for it has grown year on year. It works as a bearer asset and a settlement network at once, and anyone can confirm the entire supply from an ordinary computer without asking permission from a soul.

Set that against the rand. The SARB publishes its money supply figures and I do not doubt them, but you cannot verify in real time how many rand exist or are being created this quarter. You trust the institution to manage the currency responsibly and absorb the cost when the pressure wins. With Bitcoin there is no institution to trust in the first place. There is code and mathematics and a network of machines that enforce both, while the owners never have to trust each other. I make that case in full in Bitcoin as sound money, and the honest hedge argument around it in Bitcoin as a hedge.

The comparison with gold is worth pausing on, because gold has done this job for centuries and deserves the respect. Gold stayed sound for so long because it was the hardest thing on earth to produce, its supply creeping up under 2% a year no matter how badly anyone wanted more. But gold's supply still answers to price. When the metal runs, miners reopen marginal shafts and refine poorer ore, and eventually new supply arrives to meet the enthusiasm. Bitcoin's supply answers to nothing. More miners and faster machines cannot conjure one extra coin, because the protocol simply makes the puzzle harder so the schedule holds. The work is real and the electricity is real, and none of it moves the total by a single satoshi.

What to do with this if you save in rand

The conclusion is not that you should hold no rand. You earn in rand and spend in rand, and your emergency fund belongs in rand, boring and instantly reachable. The portion of your wealth meant to sit untouched for a decade is a different question, and it deserves an instrument that cannot be quietly expanded while it waits.

Bitcoin will not behave like a fixed deposit on the way there. A small annual inflation of 2% in rich economies, and the 5% or 6% closer to South Africa's long-run experience, taxes cash every year without ever sending you a bill. At 6% your purchasing power halves in about twelve years.

Bitcoin's supply schedule cannot be redesigned to serve a budget cycle, which is exactly the property fiat lacks, and it is why I keep it at the centre of every client conversation even though its price can fall 30% in a quarter and has done so more than once. That volatility is the cost of a young asset finding its level in an open market with no central bank smoothing the ride, which is why I insist on a five to ten year horizon with everyone I work with.

In practice I structure that as steady monthly buying from R1,000 a month, moving clients into genuine self-custody from R10,000 where the keys sit in their hands rather than on an exchange, with larger holdings going into the Vault so no single device or burglary can move the coins. If the rand doing nothing in your account is the thing keeping you up, start a monthly Bitcoin savings plan and let time and a fixed supply do the work the currency cannot.