Positive Feedback Loop of Maldivian Governance

10 Jun, 2025

In engineering, a system is defined by its boundary conditions and the objective function it has been built to optimise. When I analyse the propulsion system of an aircraft, I know its objective function without ambiguity: maximise thrust, subject to constraints on fuel flow, inter-turbine temperature, and structural integrity. Every component, from the fuel nozzle to the turbine blade, serves that single, unified purpose, and if it doesn't, it gets replaced. Having spent a career trained to read the world this way, through the laws of thermodynamics and mechanics, it was a fairly natural next step to point the same analytical framework at the country I live in. Political philosophy, in the Rousseau or Locke tradition, describes the State as a social contract designed to optimise something like the Public Good. So the question worth asking is whether the Maldivian political system, observed as an input-output machine from 2008 to the present, actually fits that objective function. If it did, the inputs, taxes, loans, foreign aid, would correlate positively with the outputs, quality of life, equitable distribution, efficient services. What we observe instead is closer to a negative correlation, a system leaking energy at every stage of conversion. That forces the honest engineering question: is the machine broken, or is it working exactly as designed, just for a different objective function than the one we assumed?

To answer that we have to abandon a habit that political philosophy indulges more than it should, which is treating the government or the party as a single thinking entity. We say things like "the government wants to improve healthcare," and mathematically that's a fallacy of composition, because the government does not think. Only individuals think. Public Choice Theory, which occupied a fair amount of my spare reading in an earlier phase of curiosity about philosophy and history, gives the more accurate model. Let $S$ be the State, not a unitary actor but a set of $n$ individual actors, politicians, bureaucrats, voters:

$S = {i_{1}, i_{2}, . . ., i_{n}}$

Each actor $i$ is not a benevolent altruist but a rational utility maximiser, operating on a utility function that weighs Power ( $P$ ), Wealth ( $W$ ), and Status ( $S$ ):

$M a x i m i z e U_{i} (P, W, S)$

The engineering flaw baked into Maldivian "Democracy v1.0" since 2008 is the assumption that the politician's utility function is aligned with the public's:

$U_{p o l} \neq U_{p u b}$

In a thermodynamic system, when the internal energy of the components is directed against the system's intended work, you get heat instead of motion, waste instead of output. Model the politician's utility function honestly and you find it pointed, not toward public service, but toward re-election and rent extraction, and the naive model of governance, maximising public welfare $W_{p u b}$ subject to a budget constraint, gets replaced by the real one:

$M a x i m i z e : f (V, R)$ $S u b j e c t t o : G_{c u r r e n t} \geq G_{m i n}$

where $V$ is vote share, $R$ is private rent, and $G_{m i n}$ is the bare minimum governance required to avoid revolution or state collapse. As long as STELCO or Fenaka keep the lights on most of the time and WAMCO eventually collects the waste, that constraint is satisfied. Whatever revenue remains after clearing $G_{m i n}$ gets diverted into patronage jobs that maximise $V$ and into corruption or inflated contracts that maximise $R$ , rather than returned to the public as savings.

In control theory, a black box is a system where you can only observe inputs and outputs, not the internal mechanism. Feed the Maldivian black box democracy, periodic elections, separation of powers, and capital, loans from India and China, tourism tax receipts, and what comes out is unstable coalitions, recurring debt crises, and unfinished projects. The naive observer's response is that we just need better people inside the box. My honest answer, and I accept I may be wrong here, is that the internal mechanism is designed to produce exactly this output. The mechanism is interest group politics. Small, concentrated groups, tycoons, business magnates, gang leaders, have a strong incentive to organise because the potential payoff per member is enormous. The general public, scattered across small islands, is diffuse and hard to organise because the cost of organising is high relative to what any individual stands to gain. The system, inevitably, tilts toward the concentrated interest.

None of what we call the failures of the system, the corruption, the inefficiency, the gridlock, are accidents. They are its equilibrium state. A ball rolling into a valley comes to rest at the bottom, and that resting point is its equilibrium, not a malfunction. The Maldivian political system has found its equilibrium in a state of high-cost clientelism.

The politician is not corrupt in a moral sense, he is rational in a mathematical one. He is responding to the incentives we coded into the system. If I design an engine to burn fuel and produce noise instead of thrust, I cannot blame the engine for being loud.

The useful move, then, is to stop analysing Maldivian politics as a contest of good against evil and start analysing it as an engineering problem of incentives against constraints.

In fluid dynamics and thermodynamics, the solution to any differential equation is entirely determined by its boundary conditions, and if you set the initial parameters wrong, no amount of correction downstream saves the system from diverging, a lesson I've learned the hard way in contexts with rather less at stake than a constitution. The ratification of the 2008 Constitution was our step function, an attempt to move instantaneously from $t = 0$ , semi-autocracy, to $t = 1$ , liberal democracy. We imported a piece of software, the Western separation-of-powers architecture, and ran it on hardware it was never built for, a small-island patronage society. The economist Anthony Downs, in An Economic Theory of Democracy, supplies the model for why this particular boundary condition failed the way it did. Downs treats political parties as firms competing in a marketplace, firms that maximise profit against consumers who maximise utility. Translate that to the Maldivian political market established in 2008 and the parties, MDP, PPM, PNC, JP, whichever coalition they happen to be wearing this decade, maximise votes, and the voters maximise rents, jobs, cash, handouts. We assumed MDP and PPM/PNC were ideological vehicles, one carrying liberal democracy and the other nationalism or development. Under Public Choice Theory that's false. Ideology is the marketing packaging used to differentiate an identical product, which is access to state resources.

The clearest numerical proof of this sits in the arithmetic of a parliamentary campaign. A Majlis seat pays roughly MVR 80,000 a month including committee allowances, which over a five-year, sixty-month term gives a legitimate lifetime income of

$I_{l e g i t} = 80, 000 \times 60 = 4, 800, 000 MVR$

Winning a competitive constituency, however, routinely costs between MVR 3 million and MVR 10 million, sometimes considerably more in Male' or other key seats. Run that as a net present value calculation, assuming a rational actor,

$N P V = \sum_{t = 1}^{60} \frac{S_{t} + R_{t}}{(1 + r)^{t}} - C_{c a m p a i g n}$

where $S_{t}$ is the official salary, $R_{t}$ is rents from bribes, contracts, and influence peddling, and $r$ is a discount rate kept high by political instability, and the arithmetic only closes if $R_{t}$ is meaningfully greater than zero whenever campaign cost exceeds legitimate income, which it usually does once opportunity cost and inflation are counted.

It is mathematically impossible for a rational agent to spend MVR 10 million securing a job that pays MVR 4.8 million, unless the position grants access to illicit revenue exceeding the difference.

Corruption in the Majlis, on this reading, is not a moral failing of any individual member. It is a mathematical necessity generated by the boundary conditions of our campaign finance system.

Materials science gives us hysteresis, the dependence of a system's current state on its history, a magnet that stays magnetised after the field that magnetised it is removed. The 2008 Constitution tried to demagnetise the Maumoon era patronage networks and the hysteresis proved stronger than the reform. What actually happened was a shift from monopoly, one-party rule, to oligopoly, multi-party rule that behaves like the textbook cartel, colluding rather than competing. Floor-crossing and coalition-flipping are what that collusion looks like in practice, the elite political class, regardless of which colour they're currently wearing, sharing the same class interest in keeping the barriers to entry high enough that no independent challenger disrupts the arrangement. This is Robert Michels' Iron Law of Oligarchy doing exactly what it predicts: every complex organisation, however democratic at founding, drifts toward oligarchy, and the 2008 Constitution mostly formalised the rules by which our oligarchy now negotiates its spoils.

Borrowing a term from information theory, a functional democracy needs a high signal-to-noise ratio, where the signal is policy substance, fiscal policy, education reform, and the noise is rhetoric, scandal, and identity politics. Because our parties offer essentially the same economic product, debt-financed infrastructure plus public-sector employment, they have nowhere left to differentiate except on noise: sovereignty scares, religious framing, personal attacks. Once the signal-to-noise ratio drops far enough, the feedback loop that's supposed to let voters correct bad policy simply breaks, and a voter on an island can no longer tell a policy that bankrupts the country from one that builds it, because the noise of the rally drowns out the signal of the spreadsheet. The 2008 Constitution was meant to set us free, and instead it put a price tag on every lever of power, turning Parliament into something closer to a stock exchange where votes trade for cash and policy trades for campaign sponsorship. What we failed to account for is that in a small, interconnected society, checks and balances themselves become transaction costs, getting a bill passed requires a majority argument and then greasing a coalition. We bolted a Western-tested engine into a vehicle whose wiring harness was never built for it, and now we're surprised the vehicle is shaking itself apart.

State-owned enterprises are the clearest illustration of the same design flaw at a different scale. The second law of thermodynamics defines entropy as the measure of disorder in a closed system, and states that the total entropy of an isolated system never decreases, entropy is energy no longer available to do useful work. Looking across our SOEs, and honestly at the non-operational layers of my own industry too, what I see are entropy machines, organisations that absorb high-quality energy, capital, tax revenue, foreign loans, and convert it into low-quality energy, bureaucracy, redundancy, patronage. The useful work, electricity generation, fish exports, infrastructure, waste collection, is a byproduct of the actual operating objective, which is vote maximisation.

In a competitive labour market, a firm hires an additional worker only when the marginal revenue product of that worker, $M R P_{L}$ , meets or exceeds their wage:

$M R P_{L} \geq W$

If $M R P_{L} < W$ , a rational firm cuts the role or improves efficiency. Our SOEs break this equation because the hiring manager is not optimising for company profit, $π$ , but for political capital, $K_{p o l}$ . The revenue an SOE employee generates is not financial, it's the vote yield, $V_{y}$ , of that employee and their extended family, plus the network loyalty, $N_{L}$ , earned by doing a favour for a constituent. The equilibrium equation becomes

$α (V_{y} + N_{L}) \geq W_{c o s t}$

where $α$ reflects the desperate marginal value of a vote in an important constituent. Since the wage is paid from the public treasury and the benefit accrues to the party, the cost is socialised while the benefit is privatised, a textbook negative externality.

This is what confuses foreign observers, the IMF and the World Bank among them, who ask why a small power plant on an island of 500 people needs 40 staff when 4 technical staff could run it. They're evaluating it as a power station. It functions as a welfare distribution node. We don't have a comprehensive, transparent social security net in the Western sense, we have the SOEs instead. Hand a citizen MVR 8,000 a month for doing nothing and fiscal conservatives call it socialism. Give that same citizen a title, Assistant Administrative Officer at Fenaka, a desk, and MVR 8,000 a month to do effectively nothing, and we call it employment. This disguised unemployment lets the government report job-creation numbers that are mathematically fictitious, what I'd classify as system noise, economic activity without economic value.

There's a second failure mode riding on top of the first, which is crowding out. Two objects can't occupy the same space, and in the same way, the public and private sectors compete for the same finite labour pool. Because SOEs operate under soft budget constraints, they can run deficits indefinitely because the Ministry of Finance will always bail them out, they can offer wages and security the private sector cannot match, $W_{S O E} > W_{P r i v a t e}$ . Why would a young Maldivian take the high-risk, high-effort route of a private workshop or a small business when a zero-risk, low-effort, guaranteed salary is sitting at an SOE or in a political appointment? The SOEs act as a gravitational well, pulling talent and ambition out of the workforce, raising the opportunity cost of entrepreneurship, and keeping the only sector that actually generates wealth chronically underpowered.

The mechanism holding this in place is a broken principal-agent relationship. A private airline CEO answers to shareholders who demand dividends, and if the CEO fills the office with incompetent relatives, the stock drops and the CEO is fired, a tight, corrective, negative feedback loop. Our SOEs have about 450,000 shareholders in theory, the citizens, but no mechanism to fire an SOE managing director directly, because the board is appointed by the same politicians demanding the hiring. The loop here runs positive, which in control systems means destabilising, not virtuous: a party wins and stuffs the SOEs with supporters, SOE debt rises, service quality falls, blackouts and delays follow, public anger rises, the party borrows more to subsidise the SOE and suppress prices to calm the anger, and the cycle repeats from the top.

None of our SOEs are businesses in any rigorous sense. They're political boilers, and we shovel public money into them to generate the steam of political support. The inefficiency, forty people doing four people's work, isn't a management failure waiting to be fixed. It's the product. Every redundant employee is a loyal voter. You cannot cheat thermodynamics indefinitely, you cannot consume more energy than you produce forever, and eventually the boiler runs out of fuel, loans or reserves, or the pressure gets too high, a debt crisis, and the vessel ruptures. I've been watching that pressure gauge climb into the red zone for a while now.

A different failure sits with the voter, and it's worth being precise about what kind of failure it is. In an aircraft, every input I make runs on high-fidelity data, and if an indicator light malfunctions or an instrument gives a false reading, the consequences are immediate, which means the cost of ignorance in aviation is effectively infinite, I have to know the true state of the system at all times. Democracy assumes the voter operates with something like that same diligence, that the citizen, as principal, is carefully auditing the politician, as agent. The obvious question is why voters keep re-electing policies that cause inflation, why known kleptocrats get returned to office, whether the electorate is simply uninformed. As a Maldivian who has a fair amount of affection for his own people, I refuse the stupidity explanation, partly on the evidence and partly, I'll admit, to preserve my own sanity, and prefer the structural one. The concept that explains it is rational ignorance.

Acquiring accurate information about governance is expensive. Understanding a budget deficit or the true exposure created by a sovereign guarantee takes hundreds of hours of reading economics, audit reports, and cutting through rally noise, call that cost $C_{i n f o}$ , time plus cognitive effort. The probability that any single vote decides a national election is statistically indistinguishable from zero,

$P_{d e c i s i v e} \approx \frac{1}{N}$

with $N$ , the voting population, at roughly 280,000. The instrumental benefit of voting correctly, identifying the genuinely best long-term policy, is shared by the whole nation and captured by no individual voter, a straightforward public goods problem. Downs' standard rational voter equation is

$R = (P \times B) - C_{i n f o}$

and since $P \approx 0$ , the first term vanishes, leaving

$R \approx - C_{i n f o}$

For a rational individual, the expected return on becoming a policy expert is negative. Spending a hundred hours studying the national budget when your vote carries a one-in-two-hundred-and-eighty-thousand chance of mattering is not diligence, it's a poor investment.

And yet Maldivians turn out to vote in enormous numbers, often above eighty percent, which only makes sense once you notice the transaction has been altered by a new variable, private rent, $D_{r e n t}$ , the direct, excludable payoff to the individual voter: cash in an envelope, an SOE job, a waiver on a loan, a promise to pave the specific road in front of a specific house, an example I know from direct experience rather than theory. The revised voter equation becomes

$U_{t o t a l} = (P \times B_{n a t i o n a l}) - C_{i n f o} + D_{r e n t}$

and since $P \approx 0$ again, it collapses to

$U_{t o t a l} = D_{r e n t} - C_{i n f o}$

To maximise utility, the rational strategy is to minimise $C_{i n f o}$ , skip the manifesto, skip the debt figures, and maximise $D_{r e n t}$ , vote for whoever offers the most immediate, tangible reward. That's why a voter in Fuvahmulah will rationally choose the candidate promising to pave the road in front of their house over the candidate promising fiscal consolidation. The paved road is real. The fiscal consolidation is invisible.

There's a second, psychological layer here, which the economist Bryan Caplan calls rational irrationality. People derive genuine psychological utility from what they believe, and in our politics, that belief is identity rather than policy, Yellow against Pink against Blue against Red. Admitting your own party is corrupt carries a real cognitive cost, dissonance. Believing your leader is the saviour carries a real psychological benefit, comfort, tribal belonging. Since the material cost of holding a false belief is close to zero, because no single vote changes the outcome anyway, it becomes rational to indulge the bias. The ballot box functions less as a survey of truth and more as a cheerleading mechanism, and the voter is consuming the feeling of political participation rather than performing the function of political oversight.

Engineering deals constantly in time constants, a concrete structure has one measured in decades, turbulent airflow one measured in milliseconds. The average Maldivian voter is trapped in an unusually short time constant, because the economic precariousness of island life keeps their effective discount rate, $r$ , very high:

${Value}_{f u t u r e} = \frac{{Value}_{p r e s e n t}}{(1 + r)^{t}}$

If someone is struggling today to pay the electricity bill, a promise of a stable economy ten years out has a net present value of essentially zero, while five hundred rufiyaa in hand right now carries close to infinite relative value. Prioritising the immediate handout over long-term stability reads as hyper-rational behaviour given a genuine survival constraint, not as shortsightedness in the pejorative sense. The voter is solving for $t = 0$ because they cannot afford to solve for $t = 10$ .

None of this is a failure of intelligence. It's calculated efficiency in an environment where accurate information is expensive and political promises are rarely kept, which leaves the only reliably real thing in politics as whatever you can hold in your hand right now. When I watch a man sell his vote, I don't see a corruption of democracy, I see a market actor reading price signals correctly, someone who has identified his vote as a liquid asset and is cashing it out for maximum immediate value. The tragedy is that this behaviour, individually rational, aggregates into a system that is collectively suicidal, incapable of long-term planning because every individual choice is optimised for the short term. Picture something close to five hundred thousand engineers ignoring the structural cracks in a shared fuselage because everyone is too busy bidding on who gets to hold the rivets.

The same logic explains our relationship with infrastructure. In a vacuum, the value of a structure is its utility, traffic flow, load-bearing capacity, economic internal rate of return, and the only honest question is whether the volume of demand justifies the capital expenditure. In the political ecosystem we actually operate in, infrastructure isn't built to solve logistical problems, it's built to solve legitimacy problems. Since 2008, and accelerating sharply after 2013, we've entered an edifice-complex phase, pork-barrel politics scaled up to the level of mega-projects, and the proliferation of loss-making regional airports and large reclamation schemes are best understood as mechanisms of wealth transfer and political signalling rather than transport or housing policy.

The bias toward capital expenditure over operating expenditure has a clean Public Choice explanation, visibility and rent extraction. A new twelve-storey building is a visible, photographable signal of development, ideal for a campaign brochure, while preventive maintenance, fixed pipes, trained nurses and teachers, is invisible, you can't cut a ribbon on a repaired pipe. It's also mathematically easier to extract a kickback from a single ten-million-dollar construction contract than from thousands of small operational transactions, and a large project creates enough complexity that a ten to twenty percent cost overrun disappears into "engineering necessities." The system is biased, structurally, toward high-CapEx, low-utility projects, optimised for the groundbreaking ceremony rather than the life-cycle cost.

The policy of putting an airport within twenty minutes of every inhabited island is the clearest example in my own field, an engineering absurdity that is simultaneously a political masterstroke. Run the unit economics: a catchment of roughly 2,000 residents, one or two subsidised turboprop flights a day, minimal revenue, and costs covering runway maintenance, rescue and firefighting cover, security, and a staff roster that often exceeds fifty. The net present value is negative, and no rational private investor would fund it. The political net present value, though, is strongly positive: construction contracts flow to party-aligned firms during the campaign-finance phase, the operating phase creates a hundred new SOE jobs on the island, and the symbolism of feeling "connected" and "modern" creates a debt of gratitude toward whoever cut the ribbon. What we're building, in effect, are Potemkin airports, structures that look like aviation infrastructure and function as employment schemes.

Land reclamation takes this further, because it's the one form of rent-seeking that creates the asset from nothing. Most countries treat land as finite. Ours can manufacture it, which hands the executive branch something close to the power of a creator, real estate to sell or lease and social housing to distribute as a lottery prize to the rationally ignorant voter. I think the distribution of social housing flats is the single most powerful currency in Maldivian politics, and I want to be clear that this isn't a claim that the housing need is fake, it's badly needed, only that the selection process runs on loyalty rather than need, and the list of recipients functions as the ledger of the patronage network.

This is also why the public tolerates white-elephant projects that visibly worsen the national debt: concentrated benefits against diffuse costs. The construction firm's ten-million-dollar contract and the island's new airport are concentrated, visible benefits, and the people who receive them will fight hard to protect the project. The two-hundred-million-dollar debt behind it is spread across the entire population and future generations, showing up as inflation and currency shortages, diffuse and effectively invisible. Because the beneficiary cares more about winning the project than the taxpayer cares about stopping it, the pork wins essentially every time, and the pattern feeds a genuinely dangerous macroeconomic loop. We borrow foreign currency to build the bridge or the airport, the construction injects rufiyaa into the local economy through wages and materials, that rufiyaa chases imported food, fuel, and electronics, demand for dollars rises, the exchange rate comes under pressure, and stabilising it requires either borrowing more dollars or starting a fresh project to inject fresh capital. We are, functionally, using infrastructure projects to pay for the present. I admire the Sinamalé Bridge as a genuine feat of structural engineering, and I still think it represents a political system that can only justify its own existence through the continuous pouring of concrete, an economy that needs a constant stream of mega-projects to feed a wealth class and employ its constituents. If the cement mixers stop, the machine seizes. We are building our way into bankruptcy one ribbon-cutting at a time.

If the rest of the system runs on friction, a ten or twenty percent inefficiency tax on government contracts, the MMPRC affair of 2014 to 2015 was a genuine singularity, the point at which the ordinary laws of the system stopped applying and the State stopped regulating public resources and started liquidating them. Public Choice Theory assumes a standard principal-agent relationship, the public as principal, owner of the islands and lagoons, the government as agent, manager of those assets on the public's behalf, obligated to lease them in a way that maximises treasury revenue,

$T = R_{m a r k e t} - C_{a d m i n}$

The MMPRC scheme rewrote that equation entirely, routing state assets through a special vehicle that bypassed the treasury and landed in private accounts,

$R_{m a r k e t} \to R_{d i s c o u n t} \to P_{p r i v a t e}$

This wasn't corruption in the ordinary sense of a kickback on a bridge contract, it was a wholesale transfer of sovereign assets, fifty-plus islands and lagoons, a finite, non-renewable form of natural capital, converted into cash for political distribution with essentially zero public utility. The scale runs to high double-digit millions of dollars in direct embezzlement and billions in long-term asset value, and what strikes me most is the efficiency of the mechanism, there was almost no leakage back into the public purse. The capture was close to absolute.

The obvious follow-up question is why the system's own negative feedback loops, the judiciary, parliament, the anti-corruption commission, failed to trigger. They failed because of regulatory capture. In a functioning system, those institutions act as circuit breakers, tripping when the voltage of corruption gets too high. In the MMPRC case, the stolen funds were used to buy the circuit breakers directly, tracing to parliament members voting for the legal amendments that facilitated the theft, to judges willing to silence dissent, and to security services enforcing the resulting status quo. The system entered a kind of auto-cannibalism, the proceeds of the crime funding the dismantling of the laws that made the crime illegal, a positive feedback loop of the most destructive kind, which is to say, corruption produces money, money produces power, power produces less regulation, and less regulation produces more corruption.

What looks like an empty lagoon to a layman is prime inventory to anyone with exposure to how these assets get financed, a resource capable of generating revenue for fifty to ninety-nine years. Leasing those lagoons at a fraction of market value to well-placed clients who then subleased them to foreign hotel chains for real profit cost the State both the immediate cash and the discounted future cash flows for generations, and running the numbers on fifty islands over fifty years against what MMPRC actually collected shows a catastrophic destruction of value, high-value, long-duration assets liquidated for low-value, short-term cash.

The most damaging legacy of the whole affair probably isn't financial, it's psychological. Safety engineering has a term for it, normalisation of deviance, a culture where an unsafe practice becomes standard simply because nothing bad happened the last time it was tried. Since 2015, the political class has learned there is no real consequence for theft at scale, and the beneficiaries remain active in politics today, crossing the floor between MDP, PPM, and PNC as convenient, with "the list" of who was involved held over their heads as leverage. The lesson the singularity taught was that if you steal big enough here, you don't go to jail, you become a kingmaker, and the old Dhivehi saying "Rihaakuru fulhi vagah nagaigen umrah jalah" captures the absurdity of that selective justice better than I can in English. MMPRC proved, as definitively as anything could, that our democracy had been hacked, that the State had stopped functioning as a commonwealth and started functioning as a private wealth extraction machine, and the singularity itself has passed while we remain inside its event horizon, the precedent that sovereign assets are personal property for the well-connected hasn't been reversed, only decentralised across the system that replaced it.

Control systems engineering distinguishes two kinds of feedback. Negative feedback is corrective, a thermostat cutting power once a room gets too hot, and it produces stability. Positive feedback is amplifying, a microphone picking up its own output from a speaker until the signal loops and screeches, and it produces instability and eventually structural failure. Looking at the political system from 2008 to now as a whole, my honest conclusion is that we are trapped in a positive feedback loop, not a liberal democracy in the Lockean sense but a clientelist oligarchy, where the relationship between State and citizen has become purely transactional. The voter doesn't ask for a twenty-year vision, they ask what can be delivered before the polls open, and the result is a hollow state, one that keeps the outward institutions of a modern republic, a parliament, independent commissions, courts, while those institutions have been emptied of their statutory purpose and refilled with a single function, rent distribution.

Institution	Statutory Purpose	Actual Function
SOE	Service Provision	Employment / Patronage
Project	Infrastructure	Kickbacks / Signalling
Vote	Voice / Oversight	Sale / Liquidation

A society without a shared idea of the public good functions as a collection of warring tribes dressed in the costumes of modern political parties.

If we wanted to predict the next election, or the one after it, we wouldn't need an opinion poll, we'd need to solve for the net present value of patronage. The probability of a voter choosing candidate $X$ is approximately

$P (V o t e_{X}) \approx \frac{I m m e d i a t e R e n t_{X}}{D i s c o u n t R a t e} + T r i b a l B i a s$

where immediate rent is the cash, job, or project promised now, the discount rate reflects the voter's desperation and rises with inflation and dollar scarcity, and tribal bias is the near-fixed variable of party identity. As the economic situation deteriorates, the discount rate approaches infinity,

${lim}_{r \to \infty} (L o n g T e r m P o l i c y) = 0$

which means that as the country gets poorer, the real cost of buying a vote actually falls, or at least it remains the only currency that reliably works, and populist leaders promising the most while delivering the least structural reform keep winning. The system, left alone, selects for fiscal irresponsibility. But positive feedback loops don't run forever, they're bounded by the physical limits of the hardware, and our hardware is the sovereign balance sheet. We're currently borrowing to fund the patronage that keeps the machine running, the debt-to-GDP ratio is already out of control, and we're close to the limit of what the credit card will bear. When external financing dries up, when we can no longer borrow dollars to prop up the rufiyaa and subsidise the SOEs, the machine seizes. That will be the hard landing, the moment the laws of economics finally impose the negative feedback our politics refused to build in, and it will look like austerity, currency devaluation, and mass unemployment.

I write all of this as a realist rather than a pessimist, because recognising the failure mode is the first real step toward fixing the machine. What we don't need is better politicians. What we need is better boundary conditions, decoupling economic livelihood from political loyalty through genuine privatisation, lowering the signal-to-noise ratio through real decentralisation of power, and lowering the cost of information enough that a voter can actually see the true price of their own vote. Until the design changes, the machine will keep doing precisely what it was built to do, which is to consume the future in order to feed the present, and I, for one, can't bring myself to respect a machine engineered to destroy itself.