Course 3 Final Review: Understanding Bitcoin’s Economic Engine

You’ve reached the end of Course 3: Bitcoin Mining & Economics, the part of Bitcoin that pays for its security and enforces its supply schedule. This review works as a set of short checks. Under each lesson summary there is one question: answer it out loud or on paper before you open the answer, and if you can’t, that lesson is worth a second look. The eight questions at the end then join the lessons together.

Part 1: How Mining Works and What It Pays

Lessons 3.1 to 3.5 follow mining from a single hash to a machine on your own electricity bill.

Lesson 3.1 — Bitcoin Mining Explained

Mining adds blocks by trial and error: machines hash a block header over and over until one result falls below a target. Miners fill blocks by fee rate, collect the subsidy plus fees, and every node checks each block itself. The difficulty adjustment keeps blocks about 10 minutes apart. Rewriting recent history with new machines would mean out-hashing the whole honest network, more than 1,000 EH/s today; the more realistic risk is pressure on the few pools that build most blocks.

Check: half the miners switch off. Why do blocks slow down, and what brings them back to 10 minutes?

Answer

The remaining machines need twice as long on average to find a winning hash, so blocks come about every 20 minutes. At the next adjustment (every 2,016 blocks) every node raises the target, making a winning hash easier to find. If the slowdown began partway through the period, that adjustment only partly makes up for it, and the one after brings block times back to about 10 minutes. After China’s 2021 ban, difficulty fell in several steps, the largest about 28%.

Lesson 3.2 — ASIC Miners

Mining hardware went from CPUs to GPUs to FPGAs to ASICs, chips that do nothing but SHA-256. Efficiency in joules per terahash (J/TH) decides which machines survive: the 2024 Antminer S21 needs 17.5 J/TH, and Bitmain’s most efficient model on sale in November 2026, the water-cooled S23 XP Hydro, 8.9 J/TH. Full-size machines need a dedicated high-power circuit (240 V in North America), ventilation and noise control. Buy from manufacturers or established resellers, and install firmware only from the manufacturer or the firmware developer’s own site.

Check: why can’t one Antminer S21 on household electricity mine profitably?

Answer

At November 2026 numbers it earns about $6.65 a day and uses 84 kWh, so it stops covering its own power at about $0.08 per kWh. Typical household electricity costs more than that: the US average is forecast at about 18 cents per kWh for 2026, and EU households paid about €0.28 per kWh in the first half of 2026. So it loses money every day, even before you count the machine’s price.

Lesson 3.3 — Mining Pools Explained

One S21 alone would find a block about once a century on average, so miners join pools that share rewards by work done. FPPS gives steady pay; PPLNS-type schemes vary with luck. Most pools hold your earnings until payout; coinbase-payout pools such as OCEAN pay you inside the block once your earnings pass the pool’s payout threshold, and hold smaller balances until then. In most pools the operator builds the block template, and the top three pools found about 63% of blocks in the month to 5 November 2026. That is why Stratum V2 and DATUM, which let miners choose their own transactions, matter.

Check: a pool doesn’t raise your expected earnings. So why does almost every small miner join one?

Answer

Because of variance. Mining alone, one S21 has about a 1% chance of finding a block in a year; in a pool it is credited about 8,750 satoshis a day, minus the pool’s fee. The average is the same either way, but the pool turns a lottery ticket into income you can plan around while the electricity bill arrives every month.

Lesson 3.4 — Bitcoin Mining Profitability

Profitability depends on electricity price, hardware efficiency, bitcoin’s price and difficulty. Hashprice combines the last two into what one unit of hashing power earns per day, and break-even electricity = hashprice ÷ (J/TH × 24 ÷ 1,000). Difficulty was roughly flat over the past year but has risen steeply over Bitcoin’s history, so a careful miner models both cases.

Check: hashprice is $0.033 per TH/s per day. What is the break-even electricity price for a 13.5 J/TH machine?

Answer

13.5 × 24 ÷ 1,000 = 0.324 kWh per TH/s per day, and $0.033 ÷ 0.324 ≈ $0.10 per kWh. Above that, the machine costs more in electricity than it earns.

Lesson 3.5 — Home Mining and Mining Scams

Home mining gives a newcomer three useful things to do: run a small miner to learn how it works, enter the solo-mining lottery, or use a miner as an electric heater. Only full-size machines need a 240 V circuit; for any miner, divide watts by volts (or use the label’s input current if it is higher) and keep the amps under 80% of the breaker’s rating. At November 2026 hashrate, a Bitaxe (about 1.2 TH/s) has about a 1 in 16,400 chance of a block in a year, and every block is a fresh draw. A miner heats a room exactly like an electric heater of the same wattage and pays a little back, which beats plain electric heat in winter but loses to a heat pump. Beginners lose real money on offers that sell mining income instead of hardware: cloud-mining contracts, phone “mining” apps and fake hosting deals.

Check: a Bitaxe on a solo pool, or the same Bitaxe in a regular pool: which earns more on average, and how does that compare with its electricity bill?

Answer

On average they earn the same, about $15 a year at November 2026 numbers before pool fees: about 50 satoshis a day in a regular pool, or a tiny chance at a whole block when mining solo. Its roughly 18 W costs about $28 a year at 18.2 cents per kWh, the average US household price forecast for 2026. Either way the electricity costs more than the device earns on average, so choose by what you enjoy and treat the bill as the price of the ticket.

Part 2: Supply and Security

Lessons 3.6 and 3.7 look at what pays miners over the long run: a subsidy that halves on schedule, and fees that follow no schedule at all.

Lesson 3.6 — Bitcoin Halving Explained

Every 210,000 blocks the subsidy halves, which caps supply just under 21 million; the subsidy reaches zero at the 33rd halving, around 2140. Nodes enforce this themselves; there is no vote. On the day of each halving, hashprice falls by about half; the least efficient machines switch off, and difficulty falls at the next adjustment. Each halving so far was followed by a price rise, the latest much smaller, and whether the halving caused those rises is debated. The next one is at block 1,050,000, expected around early-to-mid 2028.

Check: before 2024 the network issued about 900 BTC a day; after, about 450. Why doesn’t that tell you how much miners sell?

Answer

Those are new coins created, not coins sold. Miners sell some of their new coins, and sometimes older ones, depending on their costs and plans. The halving halves new supply; what it does to selling, and to price, is a separate question.

Lesson 3.7 — Transaction Fees and the Security Budget

A fee is a bid for block space, priced in satoshis per virtual byte (sat/vB); the amount sent plays no part. Transactions wait in the mempool, and miners fill each block with the highest fee rates first. The subsidy plus fees, the total paid to miners, is called the security budget, and fees were well under 1% of it in November 2026. After the 2028 halving, paying miners the same number of bitcoin would take about 80 times today’s fees; with fees unchanged, keeping their pay the same in dollars would take a doubling of the price. Whether fees can carry Bitcoin’s security, and fixes such as a permanent small subsidy (tail emission), are open debates.

Check: block 974,789 was full, yet its 7,316 transactions paid about $107 in fees in total. How can a full block be so cheap?

Answer

A place in a block costs what the cheapest transaction that still fits is paying. That morning almost the whole queue bid between 0.1 and 0.2 sat/vB, close to the 0.1 sat/vB floor that Bitcoin Core nodes pass on by default, so the block filled up with low bids. The price climbs only when more than a block’s worth of transactions bid well above the floor, so that people who need the next block have to outbid each other.

Part 3: Buying and Holding

Lessons 3.8 and 3.9 turn to bitcoin as something you buy and hold. What buyers are willing to pay sets bitcoin’s price, and the price decides what the bitcoin miners earn is worth.

Lesson 3.8 — Bitcoin ETFs Explained

Spot Bitcoin ETFs launched in the US in January 2024 and give price exposure through a brokerage account. GBTC went from a closed-end trust with large premiums and discounts to an ETF. The fund’s bitcoin is generally not insured by the fund, and US protections like SIPC cover broker failure, not losses in value. ETF buying and selling moves the price, but it gives the funds no say over Bitcoin’s rules.

Check: what does a spot ETF hold that a futures ETF doesn’t?

Answer

Bitcoin itself, kept by a custodian. A futures ETF holds contracts that expire and must be replaced, which costs money when later contracts are dearer (contango), so it tends to trail bitcoin’s price.

Lesson 3.9 — Dollar Cost Averaging Bitcoin

Dollar cost averaging means buying a fixed amount on a schedule. It removes timing decisions and lowers your average cost by buying more when prices are low. It does not prevent losses while prices fall: a DCA started at the October 2025 top was about 2% down in November 2026, thirteen months later, against about 40% for a lump sum. Withdraw to self-custody in batches and keep records for taxes.

Check: why does a fixed weekly amount give an average cost below the average price?

Answer

Because the same amount buys more satoshis when the price is low and fewer when it is high. The cheap purchases carry more weight in your total, which pulls your cost per bitcoin below the simple average of the prices.

Key Economic Concepts from Course 3

Beyond the nine lesson checks above, make sure you can explain these in your own words:

• Fee rate: what a transaction pays per unit of block space (sat/vB); miners rank transactions by it, and a full block can still be cheap when nearly everyone bids the minimum.
• Hashrate and hashprice: total guessing power (about 1,035 EH/s in November 2026), and what one unit of it earns per day.
• Expected value and variance: a pool leaves your average earnings the same and smooths out the swings; solo mining pays the same average as one rare, large payment.
• Template control: who chooses the transactions in a block, and why that matters for censorship.
• Node-enforced rules: every full node checks every block, and the rules change only when users choose to run new software. Tightening them is a soft fork (Taproot in 2021 was one): nodes that don’t upgrade still accept the new blocks. Loosening them, for example to raise the 21 million cap, is a hard fork: nodes that don’t upgrade reject the new blocks, and the chain splits. No vote can make a node accept a block that breaks the rules it runs (see who enforces Bitcoin’s rules).
• Custody: who holds your money before it is yours to move: a pool, an ETF’s custodian, an exchange, a cloud-mining company, or you.
• Stock-to-flow: existing supply divided by new supply; a way to think about scarcity, with a mixed record as a price forecast.
• Miner capitulation: unprofitable miners switching off after a halving or price fall, until difficulty adjusts down.
• Security budget: the total paid to miners, subsidy plus fees. Fees were under 1% of it in late 2026. Whether fees can eventually pay for Bitcoin’s security as the subsidy shrinks is the course’s biggest open question.

How These Concepts Connect

It helps to separate what the code guarantees from what people argue about.

Mechanics (set by the rules, certain): the subsidy halves every 210,000 blocks; difficulty adjusts every 2,016 blocks to keep blocks near 10 minutes; nodes reject blocks that break the rules; miners are paid the subsidy plus fees, and no more.

Economic feedback (well observed): miners’ income in dollars depends on bitcoin’s price. When income rises, more machines switch on, hashrate and difficulty rise, and each machine earns less. When income falls, the least efficient machines switch off, difficulty falls, and the rest earn a little more. Hardware efficiency keeps improving, so older machines keep being pushed out.

Hypotheses (debated): that halvings drive price up by cutting new supply; that ETF demand and steady buying such as DCA create lasting upward pressure; that a higher price automatically means more security; that demand for block space will grow enough for fees to replace the subsidy. Each has evidence on its side and serious critics. Demand from ETFs and individual buyers helps set bitcoin’s price, and the price decides what the bitcoin miners earn is worth, but no rule guarantees how much demand there will be.

For how bitcoin compares as a store of value, see Bitcoin vs. gold, stocks and fiat.

What’s Next? Course 4: Lightning Network & Bitcoin Nodes

Course 3 covered how Bitcoin’s base layer pays for itself. Course 4: Lightning Network & Bitcoin Nodes moves to using and checking Bitcoin yourself. You’ll learn:

• The Lightning Network: how payment channels make payments fast and cheap on top of Bitcoin
• Channels and routing: how channels open, move money and close
• Lightning wallets and payments in practice: tools for sending and receiving
• Running a Bitcoin node: why it matters for verification and privacy, how node platforms compare, and how to set one up
• Running a Lightning node: what it takes to run your own
• Backups and watchtowers: how to keep the bitcoin in your channels safe

Course 3 explained why every node checks every block. Course 4 shows you how to run one yourself.

Check Your Understanding

1. Strong ETF buying pushes bitcoin’s price up 50%, and it stays there. Over the next few months, what happens to each machine’s earnings, to the security budget and to the supply schedule?

Answer

At first each machine earns 50% more in dollars, because hashprice moves with the price. That makes more machines worth running, so hashrate rises and, at each adjustment, difficulty rises too. Each machine’s share of the roughly 450 BTC issued per day shrinks, pulling its earnings back toward the cost of electricity for the typical miner. The security budget is 50% larger in dollars, because the same subsidy and fees are worth more. The supply schedule doesn’t move at all: the funds’ buying changes neither the subsidy nor the 21 million cap, and gives them no say over the rules every node enforces.

2. It is the day of the 2028 halving, and nothing else has changed. What happens to hashprice, to an Antminer S21’s break-even electricity price, and to the fees’ share of miner revenue?

Answer

Almost all of hashprice comes from the subsidy, so it falls by about half, from about $0.033 to about $0.017 per TH/s per day. The S21’s break-even price falls with it, from about $0.08 to about $0.04 per kWh, so many machines stop covering their electricity overnight. Fees don’t halve, so their share of miner revenue roughly doubles without fees growing at all. Then the least efficient machines switch off, difficulty falls at the next adjustment, and the machines still running earn a little more.

3. Someone says “the halving makes the price go up, so mining always becomes profitable again.” Which parts are mechanics and which are hypotheses?

Answer

Mechanics: the halving cuts the subsidy in half on schedule, and if miners leave, difficulty falls so the rest earn more per machine. Hypotheses: that the halving itself makes the price rise, and that a rise will come soon enough to save any particular miner. Past halvings were followed by rises, but four examples, with the latest much smaller, can’t guarantee the next.

4. Why is the security budget called the course’s biggest open question, and what answers do people give?

Answer

The subsidy halves about every four years and reaches zero around 2140; after that, fees alone pay miners to keep Bitcoin’s history expensive to rewrite. Fees were under 1% of miner revenue in November 2026, and after the 2028 halving it would take about 80 times today’s fees to pay miners the same number of bitcoin. Some expect demand for block space to grow into the role; some say a rising price buys time; some argue today’s security is more than needed; others read the falling fee share as a warning. The most debated fix, a permanent small subsidy (tail emission), would break the 21 million limit and take a hard fork. The 2026 discussion among developers ended without agreement, and nobody can settle the question today.

5. A pool can’t take coins from your wallet. So who decides which transactions go into a block, and what limits that power?

Answer

Whoever builds the block template: in most pools, the operator. To collect the most fees, it fills the block with the highest fee rates first, but each pool sets its own policy, and in the month to 5 November 2026 the top three pools found about 63% of blocks. So a few operators could leave out transactions they don’t want. Three things limit that power. A skipped transaction waits for a block built by someone else, and only a majority could refuse to build on such blocks. Miners can switch pools in minutes. And Stratum V2, DATUM or a solo pool run on your own node let miners build their own templates. Every node still checks every block, so no template builder can include an invalid transaction or pay itself more than the subsidy plus fees.

6. A website sells “hashrate”: 100 TH/s for a year, with a guaranteed $10 a day. Which number lets you test the offer, and what does the test show?

Answer

Hashprice. At about $0.033 per TH/s per day (November 2026), 100 TH/s earns about $3.30 a day before electricity and pool fees, so $10 a day is three times what that hashing power can produce. Real mining income moves every day with price, hashrate and fees, so a guaranteed amount is the warning sign, and a return above what the machines earn has to be paid from somewhere, usually new customers’ money. The same number tests a seller’s payback chart: multiply hashprice by the machine’s TH/s, subtract the electricity at your own all-in price, and remember that the 2028 halving will cut the revenue side roughly in half.

7. Across this course, bitcoin owed to you can sit with someone else in four places. Name them, and the step that reduces the risk in each.

Answer

• A mining pool’s unpaid balance: keep the payout threshold low enough that little sits with the pool, but not so low that every payout is a tiny coin; or use a pool that pays you inside the block once you pass its threshold.
• An ETF: you own shares, a claim on a fund whose bitcoin a custodian holds, generally without insurance from the fund. Read the custody risks in the prospectus, or hold some bitcoin yourself.
• An exchange during a DCA plan: withdraw to a wallet whose keys you hold, in batches.
• Cloud mining or a hosting deal: pay for hardware you can check, never for promised income, and have your machine hash to your own pool account or address.

8. Why can withdrawing $10 of bitcoin every week, or setting a tiny pool payout threshold, cost you later? What does that have to do with how miners choose transactions?

Answer

Each withdrawal or payout becomes a separate coin (a UTXO) in your wallet, and every coin you spend adds about 68 vB to the transaction if it sits in a native SegWit (bc1q) address. Miners rank transactions by fee rate, the fee per vB, and ignore the amount, so a payment built from ten small coins costs several times as much in fees as the same payment from one large coin. When fees rise, a small coin can cost a large share of its value to spend. Withdraw in batches and choose a sensible payout threshold. If you combine small coins into one while fees are low, remember that doing so links them publicly.