I often find myself in conversation with supporters of Modern Monetary Theory, and for a long time I was mystified by what they are saying. This is the MMT I meet in argument and online, not the academic literature, which is more careful and more divided than any single account allows.
Over time I have come to think of MMT as three layers: a small set of accounting identities, a set of mechanisms, and a set of normative principles. Some are uncontroversial. Others are either inconsistent with the rest of MMT or simply impractical. What is most evident is that the identities do not establish the mechanisms, and the mechanisms do not establish the principles. If anything the arrow runs the other way: the mechanisms flow from the principles, and the identities are recruited afterward to make the whole look derived.
Three of the identities are mainstream economics; nothing new there. MMT adds a fourth claim and presents it as though it too were an identity, but it is not. It is the first of the mechanisms, an institutional design proposal wearing accounting's clothes. Of the six mechanisms, four rest on a technically true core the mainstream accepts but add reasoning or consequences it does not accept; the other two are institutional proposals the mainstream rejects outright. None of the principles are mainstream at all.
MMT did get one thing right, and it is worth saying so at the outset. A government that issues its own free-floating currency is not a household. It cannot be forced into nominal default on its own-currency debt, and even mainstream central banks now accept as much. So this is not the tired complaint that printing money causes inflation. The disagreement is about everything the theory builds on top of that one sound insight. I will take the layers in the order MMT usually presents them, from the arithmetic up to the politics, because that is the order you meet them in argument.
The Identities
The foundation, as presented, is a handful of identities, all variations on one idea: every flow of spending is someone's income, and every financial asset is someone's liability, so the books must balance. (A house is not a financial asset; it is real wealth, owed by no one. The netting is about financial claims, not wealth as such.)
The sectoral balances identity. Split the economy into three parts, government, the domestic private sector, and the foreign sector, and their financial balances must sum to zero. If one sector runs a surplus, the others together run a matching deficit. The practical reading: a government surplus must be mirrored by a deficit elsewhere. Run a surplus while the country runs a current account deficit, and the private sector must draw down its savings to close the arithmetic. True by construction.
The two-sector compression. Collapse the private and foreign sectors into one "non-government" bucket and you get the most-quoted form: the government's deficit equals the non-government surplus, to the penny. This is the line you actually meet in argument, usually delivered as though it revealed something. It is the first identity at lower resolution. True by definition.
The national accounting identity. Output equals consumption plus investment plus government spending plus net exports. Rearranged, it yields the saving-investment identity, the same sectoral story from the income side. MMT inherited this and chose to emphasise the sectoral reading. It is the GDP equation reframed.
$Y = C + I + G + NX$ (GDP from Expenditure)
$Y = C + S + T$ (GDP from Income)
$C + I + G + NX = C + S +T$ (Equate the two equations)
All three of the above identities are unarguably true. Mainstream economics relies on them just as much as MMT; basic accounting consistency is a requirement for any serious framework, not a feature of this one. Same identities, different conclusions: these identities do not determine MMT. They are silent on whether a deficit is wise, whether spending causes inflation, or whether the currency holds its value. You can grant all three and concede nothing. The weight of the theory is borne elsewhere.
There is a fourth claim MMT presents as though it were an identity: that the consolidated government and central bank, as monopoly issuer of the currency, creates money when it spends and destroys it when it taxes, and so is not revenue-constrained like a household. That is not an identity. It is a proposal about how to wire the state, and it belongs in the next layer.
The Mechanisms
Here MMT becomes a theory you can argue with. The pattern is the key to the whole assessment: where MMT agrees with the mainstream it is sound, and where it departs it is soft. The distinctive content and the weak content are the same content.
The government can issue whatever money it needs. This is the claim that masquerades as a fourth identity. Seen plainly it is the first mechanism: wire the state so the government funds itself by creating the currency rather than raising it. How that is achieved, an accounting consolidation, a bank that accommodates the treasury, large-scale debt purchases, is implementation. The mechanism is the capability, and everything else in this layer presupposes it.
The crucial point is that this is not axiomatic. The above identities are true by definition, in every economy, under every government. This capability is a decision that a government gives effect to, an architectural choice, and architectures vary. Whether the underlying capacity exists is beside the point; a sovereign issuer has it. What MMT relies on is that the capacity is institutionally available and exercised, and that is a built arrangement, contingent and reversible, which is exactly why it is a mechanism and not an identity. The United States bars the Federal Reserve from buying debt directly from the Treasury. The eurozone treaties prohibit monetary financing outright. Australia reaches the same end by agreement rather than statute: the Treasury and the Reserve Bank simply maintain that deficits are funded in the market, not by the central bank. Some economies sit closer to the MMT wiring than others, and that variation is itself the proof that the capability is contingent, not necessary. It is a constraint built deliberately into the architecture of the state, and in Australia's case it holds only so long as the Treasury and the Reserve Bank choose to keep it, exactly the self-restraint the rest of this essay will question. Where a state enables this capacity, the nominal claim holds: the government can issue the funds it needs and cannot be forced into default. What that does and does not secure is the subject of the rest of this essay.
Suppose Australia wired the state this way tomorrow. The mechanism would deliver what it promises: the government could meet any own-currency obligation by crediting accounts, and so could not become insolvent in its own currency. Granted. But notice what that builds and what it does not. It guarantees the nominal payment can always be made. It guarantees nothing about whether the payment is costless, whether the currency holds its value, or whether the spending bids up prices or sinks the exchange rate. The mechanism secures the ability to print the numbers. It says nothing about what the numbers are worth. The financing constraint was never the binding one. The real constraint, that the government can create money but not the labour, energy, housing, and imports the money commands, is untouched.
Inflation comes from spending outrunning capacity. Below full employment, extra spending mobilises idle resources and raises output; past it, it only bids up prices. This is presented as an MMT insight, but it is the output gap restated in words, the relationship the Phillips curve describes. MMT rejects the NAIRU framing and would site the price anchor in a job guarantee rather than a pool of unemployed, which is a real disagreement, not mere wording. But the proximate cause is diagnosed identically: spending past real capacity. The mechanism is common ground, so the disagreement cannot live here.
Fiscal policy alone can target the output gap. No economist denies fiscal policy affects inflation; that half is uncontested mainstream. The big call is the next step: that fiscal policy can do the whole job, and the independent central bank can go. That half is the MMT departure, and it leans on the first mechanism. With solvency no longer a worry, a single instrument suffices: spend more and tax less below capacity, spend less and tax more above it. One target, one instrument. At the technical level this is coherent. But we built independent monetary policy for a reason, precisely because governments proved bad at the unpopular decision. The mechanism works on paper. It assumes the instrument can be moved symmetrically and promptly, tightened into a boom as readily as loosened into a slump, and that is an assumption about politics, not economics. I will come back to it. Here the mainstream and MMT sit side by side: the demand effect is shared, the single-instrument sufficiency is the addition, and the addition is the soft part.
Permanent near-zero interest rates. If fiscal policy controls inflation, the interest rate has no work left, so it can sit at or near zero. Domestically this is deliverable: a monopoly issuer can pin the overnight rate and hold it. Even at home it carries a cost, stripping the price of money out as a signal, encouraging leverage, inflating assets, which is much of the post-2008 story. But the fatal objection is external, and it earns its own section below. In brief: the authorities set the policy rate, but the rate a foreign lender demands to hold the currency is a different price the government does not control. Hold the rate at zero while deficits make depreciation likely, and foreigners leave; the adjustment falls on the currency. The rate is not a lever the government owns. It is a permission the market grants, and only while the market believes the inflation outlook justifies it.
Grant MMT its very best case. Hand the government a perfect fiscal thermostat: automatic, immediate tax rises that fire the instant inflation lifts and ease the moment it falls, believed by everyone. This is more discipline than any legislature has shown, and granting it disposes of every easy objection at once, the lag, the politics, the credibility problem. In a closed economy it works: demand held at capacity, inflation at target, the rate parked at zero forever. That is the MMT ideal at its most attractive, and in a world that does not trade, it is coherent.
The open economy breaks it anyway, and now the failure is clean, because discipline is no longer the missing piece. Imported inflation comes from the exchange rate, set by foreigners weighing yield against expected depreciation. The thermostat reaches neither. It acts on domestic demand, while the price that is moving is set abroad, and it says nothing to the foreign holder, who holds the currency for return, not to pay the ATO. Pin the rate at zero and that holder has no reason to stay; the currency falls; imported prices rise however hard the thermostat squeezes at home. So you keep taxing, using a domestic tool against an external price it cannot touch. The loop does not converge: each round crushes domestic demand deeper and leaves the cause, the unattractive currency, where it was. To offset imported inflation through the only channel the thermostat owns, you drive the economy below capacity, into the slump the theory exists to prevent, and the currency may still be falling. The instrument that would defend it, by speaking to the foreign holder's yield, is the interest rate, which the ideal has just retired. The discipline was never the missing piece. The retired instrument was.
There is a further cost, and it lands even in the closed economy. Consider who bears the tightening. Today, when the economy overheats, the central bank lifts the rate, and the pain comes through the price of borrowing, which you can see coming, fix against, refinance around, and your bank can model. Under MMT the rate is pinned, so tightening comes through the tax bill instead. To cool a boom, the thermostat raises taxes, and every household's after-tax income becomes the adjustment variable. There is no longer any certainty a household can meet its mortgage from one year to the next, because the income it services the loan from is now what policy deliberately varies. The whole credit system is built on income being the stable input; serviceability tests and the borrower's own budgeting assume take-home pay is steady and the rate is the thing that moves. MMT inverts this, fixing the rate and varying the income. Cyclical risk shifts to where credit can least bear it: a rate rise is visible and hedgeable, a tax surprise is neither, and it hits every borrower at once. Lending gets dearer and more conservative, or mispriced and more fragile. The cost of tightening did not vanish. It moved off the visible interest rate and onto household income, the collateral the banking system rests on, and surfaces as systemic fragility.
Taxes create demand for the currency. Taxes do not fund spending, MMT says; they create demand for the currency and free up real resources. The resource-freeing half is sound: to take real resources without inflation, the government must withdraw private spending power, and tax does that. The currency-demand half overreaches. In its strong form, that money has value fundamentally because the state demands it back in tax, it is a partial truth sold as a complete theory. A tax liability explains why an Australian accepts Australian dollars; it says nothing about what those dollars are worth, and worth is the question. The obligation reaches the shopkeeper in Sydney and stops dead at the factory in Shenzhen and the mill in Dhaka, who owe the ATO nothing and will part with my next computer or my clothes only for what the dollar converts into. With imports near a quarter of GDP, that is a quarter of what we buy whose price is set entirely outside the tax net. Hyperinflations happen inside functioning tax systems, which could not occur if tax demand sustained value. So tax demand can bootstrap a currency but cannot keep it valuable, and keeping it valuable while the government spends freely is the one thing MMT most needs.
The job guarantee. The standing offer of public work at a fixed wage to anyone who wants it, meant both to deliver full employment and to anchor prices through that wage floor. It is the softest joint in the layer, because it is asked to do two things that pull against each other. As a stabiliser it must expand in slumps and contract in booms, which needs jobs you can switch on and off, which means they cannot rely on continuity, capital, or coordination. But useful work is mostly continuity-dependent. The activities advocates reach for, aged care, environmental restoration, tutoring, are valuable precisely because they need training and permanence, exactly what a job designed to vanish in a boom cannot supply. The more useful the work, the less disposable; the more disposable, the more makeshift.
A distinction matters, because it is where the defence hides. To say work is wanted is not to say it is productive in the sense that counts for inflation. What lets an economy run hot without overheating is productivity growth, and a buffer-stock public pool is structurally poor at generating it, chosen for switchability, free of competitive pressure. So a large, permanent, low-productivity-growth sector drags on the very capacity needed to grow out of inflation. Press the wage and it gets worse. To anchor prices the wage must sit at the bottom, below award rates, or it competes with private employment and adds to the pressure it was meant to restrain. But a permanent public tier paid below the floor is exactly what organised labour exists to oppose. The wage that anchors is the wage the unions fight; the wage that buys industrial peace is too high to anchor. And once the work cannot be productive, the scheme collapses inward: what remains is the state paying an income to anyone who turns up, the "job" a fig leaf over a transfer. MMT rejects a basic income precisely because it does not anchor prices; but the job guarantee, stripped of the productivity that was meant to do the anchoring, anchors no better, and what remains is a basic income burdened with the cost of pretending to be a job. More than the dole, less than the market, permanent, a larger standing commitment than a benefit while producing less than a job, and harder to administer than a simple transfer. The one thing meant to distinguish it from a plain income floor, anchoring prices, depends on the productivity it lacks.
Notice what "unemployment is a policy choice", the claim the scheme serves, is doing. Narrowly it has content: demand-deficient unemployment is real and fiscal policy can absorb it, which the mainstream half-grants. The overreach is "choice", collapsing cyclical unemployment, which demand fixes, into structural and frictional unemployment, which it does not. Push demand hard enough to absorb all of it and you get the overheating the first mechanism warned of. And the claim is not only empirical; it is a moral charge, that mass unemployment is a cruelty the state could end and elects not to. That does not belong in this layer at all. It is a value commitment conjugated into the indicative so it reads like a finding, and it is the hinge on which the structure swings from economics into politics. The pattern repeats from one layer down, where the supposed fourth identity turned out to be the first mechanism. The structure is shingled: each layer's last member already belongs to the layer above, dressed in the costume of the layer below, and that overlap is where unearned certainty is passed up a join that should have stopped it.
The Political Program
The top layer is not economics at all. It is a set of value commitments, and there are four.
Full employment is a right. Work, or the offer of it at a living wage, is something a citizen is entitled to and the state obligated to provide.
Unemployment is imposed cruelty. Mass joblessness is a deliberate choice, a buffer stock of the jobless kept to discipline inflation, the NAIRU recast as institutionalised cruelty dressed as technical necessity.
Austerity is ideology, not economics. Solvency fears and debt limits are a constructed apparatus serving a distribution of power, restraining the state and protecting creditors.
The politics drives the economics. The accounting and the mechanisms are marshalled to show a generous, full-employment state is affordable, the motivation running from the moral commitment outward to the apparatus rather than the other way.
MMT promises an economics of upside without trade-offs. Inflation controlled. Everyone who wants work has it. Borrowing costs near zero. Each promise is attractive, and together they describe a world where nothing must be given up to get anything else. But economics is the study of opportunity cost, of what must be surrendered to obtain what you want. You cannot have your cake and eat it too, and a theory whose appeal rests on saying you can is selling the one thing economics exists to say you cannot buy.
These four are slogans, not insights. The appeal is real; the deliverability is not, because each denies a cost that does not vanish because you refuse to name it.
This is the inversion at the heart of it. In a sound structure the load runs upward: foundations, then mechanisms, then conclusions the structure earns, held most tentatively because they are reached last. MMT runs the other way. The conclusions come first and most confidently, and the foundations are gestured at behind them as if they did the supporting. The identities are not the foundation in any sense that matters. They found nothing, because a definitional truth founds nothing. The real foundation is at the top: the moral commitments came first, and the mechanisms and identities were enlisted to make a chosen politics look derived. The boring identities do not buy much precisely because they were never meant to buy anything. They were meant to make a prior conviction look purchased.
Where It Breaks: The Small Open Economy
Grant the theory its best case and it still fails for an open economy, and the failure is theoretical, not just practical. The central claim, that real capacity is the only binding constraint and one fiscal instrument can manage it, assumes the constraint binds only at domestic full employment. For an open economy there is a second door.
The government spends past capacity, or funds a current account gap, with rates at zero. Foreign holders judge the currency may well buy less in future, and sell. The currency falls. Because an open economy imports heavily, that raises the local-currency price of imports directly, and that is inflation, arriving regardless of where the domestic output gap sits. The market need not believe the government will default. It need only believe the currency will be worth less, and the belief is self-fulfilling: selling delivers the depreciation that justifies the selling.
This survives every comforting fact about Australia. All government debt is in Australian dollars, most external liabilities are in Australian dollars or hedged, and the country holds a net foreign-currency asset position, so a depreciation improves the national balance sheet rather than detonating it. That defeats the foreign-currency-debt objection completely, and offers no protection against the import-price channel. A depreciation can repair the balance sheet at the very moment it raises the cost of living. The hedging answers solvency; it is silent on purchasing power.
The escape hatches close in turn. If foreigners will not fund the deficit at zero, the central bank can monetise it. Domestically that works; a central bank is always the buyer of last resort for its own debt. But it does nothing about the external gap. It substitutes for foreign demand for the bond, not for the currency, and it removes the yield signal that would have warned of the strain. The pressure accumulates silently and loads onto the one price left unfixed, the exchange rate, more violently for being suppressed. This is the impossible trinity in MMT dress: you cannot run persistent monetised deficits, hold the rate at zero, and keep the currency stable, all at once. Pick two. MMT's confidence about setting the rate and monetising the deficit is really a choice to sacrifice the currency while talking as though all three hold. For the United States the sacrifice is small, because global dollar demand is deep. For Australia the sacrificed corner is the one it cannot afford, because a falling dollar feeds straight into import prices.
So the constraint MMT promised to escape reappears as inflation. It was never escaped, only rerouted. The framework says watch one gauge, the output gap. The open economy has two, the output gap and the currency, and the second can redline while the first reads safe. To defend the currency you need the interest rate after all, the instrument MMT retired. The second instrument it abolished turns out to be necessary because of the channel it ignored.
The Curve Comes Apart
The cleanest way to see it is as a central bank problem, because it shows up as a price you can watch. The overnight cash rate is the bank's to set. The ten-year yield is not; it is the market's required return for lending a decade, pricing expected inflation, future rates, and risk. Normally the long rate is anchored to the expected path of the short rate, so the bank can pretend it controls "the" interest rate. The pretence holds only while its credibility holds. The bank directly sets only the overnight rate; the long end it can influence only by buying.
Lose credibility and the long end disconnects. A regime that holds rates at zero while monetising deficits makes the market expect higher inflation and a weaker currency, so it demands a higher yield to hold the ten-year, whatever the cash rate. The bank holds the short end at zero and watches the long end climb, because the long end is pricing the discipline the regime abandoned. The denomination facts do not help: the bond is in local currency, so there is no default risk, and the yield rises anyway, because what is priced is debasement, not default. And the monetisation response is self-defeating: the only way to suppress the long rate is to buy the bond, which confirms the fear that lifted it.
This inverts the MMT claim. The zero rate is not imposed on the market. The overnight rate the bank sets outright; the rest of the curve it can only buy, and buying confirms the very fear that lifted the yield. The long end stays low only while the market expects low inflation; otherwise it demands a return matching its own ten-year view, and you cannot legislate that away. "We can hold rates at zero" simply means "the market believes our inflation will be low", and the only way to produce that belief is the disciplined policy MMT calls unnecessary. The claim undermines itself: it asserts freedom from a constraint whose absence depends on behaving as though the constraint were present. The freedom is real precisely to the extent you do not use it.
Budget Night
Picture the Treasurer announcing the whole project, not the cherry-picked version. Deficits funded by the central bank. The cash rate to zero and held. Solvency declared a non-issue, because a currency issuer cannot run out of its own currency. And everyone without work offered a job, paid above the dole but at the bottom of the wage structure, to deliver full employment and anchor prices.
The reaction does not wait for the policy to be tested. It begins before the speech ends. The dollar falls in offshore trading as the words are read, because the market is pricing the inflation the regime has just promised not to control with the old instruments. The ten-year gaps higher: the bank still sets the overnight rate but no longer owns the long end, which reprices to the market's own view. Importers mark up overnight. None of this needs a dollar spent or a job filled. The discipline MMT believed it had retired reasserts itself the same evening, as a market that simply declines to fund the experiment on the announced terms.
The job guarantee draws its own fire, and the wage gives it away. To anchor prices it must sit below the floor, which organised labour exists to oppose, so the industrial reaction lands on top of the market's. Lift the wage for industrial peace and it anchors nothing, an open-ended payroll paid above the dole for low-productivity work. There is no figure the Treasurer can read out that is at once low enough to anchor and high enough to announce.
So no government adopts the project whole, not because the politics is merely hard, but because announcing it is the trigger that defeats it. What reaches the world is never MMT entire, only the half that does not bind, the deficits and the latitude, with the discipline and the anchor left out. Governments take the part that gives and leave the part that takes. The full project fails at the despatch box. The partial one is just spending with the safeguards stripped, which is the caricature its defenders resent, except the caricature is what democratic politics actually selects for.
The Discipline That Lets Pigs Fly
This is the deepest irony. MMT is sold as liberation from constraint, and the serious version demands more fiscal discipline than the mainstream, not less. It asks the budget to track the output gap continuously, tightened the instant the economy overheats, with the legislature behaving as a real-time inflation-targeting machine. The mainstream never asks this of fiscal policy. It parks fiscal policy on slow solvency objectives and hands the fast, unpopular, counter-cyclical job to an insulated central bank, precisely because it assumes politicians cannot do it. MMT hands that job back to the politicians and assumes a timeliness no legislature has shown.
The argument reduces to one regularity: elected governments struggle with fiscal discipline. This is structural, not incidental. The benefits of spending are concentrated and immediate; the costs of restraint are diffuse and deferred. Tightening into an upturn means imposing visible pain on identifiable people to forestall an inflation not yet arrived, in good times when the case feels abstract. Genuine consolidation almost always needs a crisis to supply cover or an external enforcer. And the struggle is asymmetric: governments manage the expansionary half easily and fail at the contractionary half, which is exactly the half MMT's inflation control needs. COVID showed it across every advanced economy at once. Every government, whatever its declared framework, loosened decisively and then could not withdraw the support on time, and the inflation followed. That is not a test of MMT, since no one ran the whole project; it is a demonstration of the asymmetry MMT must overcome and cannot.
Put it as a trilemma. The textbook version says a country with open capital markets cannot both fix its interest rate and hold its currency steady; one must give. MMT walks into a parallel one. It wants fiscal dominance, inflation control, and a fiscal authority that stays democratically responsive. It can have any two. Dominance plus inflation control needs a legislature that tightens automatically against the electorate, which is no longer responsive but the insulated technocrat MMT set out to abolish. Dominance plus responsiveness means the legislature will not tighten into a boom, and inflation control is lost. Inflation control plus responsiveness, with fiscal policy left free, is just the independent central bank restored, the orthodoxy MMT rejects. MMT insists on the corner it cannot have.
This closes every loop. The market's permission to hold the curve down depends on believing the fiscal authority will be disciplined. MMT needs that discipline to be greater than ever and lodges it in the actor least able to supply it. So the market withholds the permission, the curve disconnects, the currency falls, the inflation arrives. The framework is not undone by demanding too little restraint. It is undone by demanding superhuman restraint from the one body structurally guaranteed not to have it.
What Survives
This is no case for dismissing MMT as a fairy tale; that would be the mirror of its own error. Several things survive. The identities are sound. A currency sovereign cannot be forced into nominal default on its own-currency debt, which is why Greece could fail and Japan, with debt over twice its GDP in yen, has not. Acting decisively with large deficits in a real downturn, rather than freezing over solvency fears, is a genuine lesson. And fiscal rules really are choices with distributional consequences, not laws of physics.
What does not survive is the elevation of these into a general operating system for fiscal policy. The constraint MMT dismisses was never abolished, only relocated, from the bond market to the foreign exchange market, and made less visible and more absolute at once, because you cannot legislate foreign expectations of low inflation. The constraint is softest at the centre of the monetary system, the reserve issuer, and hardest at the periphery, the opposite of where a developing periphery's needs would want it.
So the verdict. The three identities are unarguably true and decide nothing, beyond ensuring accounting consistency. The mechanisms are robust where inherited from the mainstream and weak where original, beginning with the very first, the decision to wire the state so the government funds itself directly, which secures only the nominal result and leaves the real constraint untouched, and most exposed in a small open economy where the exchange rate defeats the single-instrument promise. The principles, which form the public face, have little foundation in either layer below. The link from identities to mechanisms to values is weak. The strength is within the layers; the weakness is in the joints, and MMT's rhetorical power comes from making the joints invisible, so that disagreement feels like innumeracy.
It is not that MMT contains no descriptions; it plainly does. The point is that the descriptions are the part that is definitional or conventional, the identities, the reserve operations, the no-default fact, true but telling you how the plumbing works, not how policy should be conducted, and the mainstream does not dispute them. The distinctive content, the part with any originality or force, begins only when MMT moves from describing the system to proposing how to reorganise it: fiscal dominance, the retirement of monetary policy, the job guarantee. That programme is the real MMT, and it must stand on its own merits, not on the accounting that precedes it and does not entail it.
And on its own merits it does not stand. It assumes a disciplined actor democratic politics does not produce, and a closed economy that a country like Australia is not. Build it whole and the market forecloses it at the despatch box; build it in part and you get the latitude without the safeguards. Either way the cost the programme was designed to abolish does not disappear. It reappears, as a falling currency, a disconnected yield curve, a brittler banking system, and the inflation the whole apparatus was meant to escape.
No comments:
Post a Comment