Can Blockchains Revolutionise Social Welfare Programmes?


Governments contemplating social welfare programmes have to wrestle with two seemingly irreconcilable problems. On one hand, voters demand social welfare programmes to take care of the poor, sick, elderly and marginalized. On the other, governments need to prevent welfare spending from ballooning extravagantly and ensure that recipients spend the money wisely. In this article, I will examine the use of blockchain technology to develop next-generation food assistance programmes.

Blockchain technology is a step up above existing schemes such as Electronic Benefits Transfer cards. Properly designed, it offers the ability to significantly reduce instances of fraud and abuse, enable quick and accurate accounting, creates the opportunity to positively shape recipients’ diets for the better, and the technology is easily transferable to other benefits schemes.

And it can do this within the next five years.

Enter NutriCoin

Today’s technology can replace EBT cards and their equivalents with a blockchain-based platform that uses internal tokens and smart contracts. For this article, let’s call this token NutriCoin.

NutriCoin requires three distinct accounts: consumer, merchant and government. Consumer accounts may only receive tokens from the government account and spend them at merchant accounts. Merchant accounts may only receive tokens from consumer accounts and trade them with the government account for fiat. The government account sends tokens to consumer accounts and receives them from merchant accounts. No other types of transactions are allowed; thus, a merchant cannot use NutriCoin to buy fresh foods, and a consumer cannot transfer his tokens to another.

Every day, a recipient uses NutriCoin to purchase groceries, meals and other essentials from authorized merchants. This transaction takes the form of a smart contract, which states that the merchant will provide goods in exchange for so many tokens. The merchant then trades the tokens with the government for fiat. This second transaction is another contract, which states the government will pay so much in fiat for so many tokens. At the end of the day (or some other block of time), the government tops off every consumer account with a fresh batch of tokens.

NutriCoin isn’t simply a next-generation food stamp. Key to its existence is restriction: it can only be used to purchase specific types of goods from licensed and vetted merchants. A drug dealer won’t be able to get a license to sell narcotics, so he won’t be able to set up a merchant account. A legitimate merchant, such as a supermarket, can only accept NutriCoin for sales of certain kinds of pre-registered food. Further, as all transactions are recorded on the blockchain, fraud can be quickly detected.

In EBT fraud, an EBT cardholder sells his card to an unscrupulous merchant (say a clerk or cashier) for a fraction of its dollar value. The clerk then uses the card to buy food items to restock the store’s shelves or key in false entries to transfer EBT funds into the store’s account. The cardholder gets free money, while the store enjoys lower operating costs or increased profit margins.

NutriCoin eliminates this by making the transaction visible. If someone suddenly splurges his entire allowance at a store, the transaction will be recorded on the blockchain, including the smart contract that details the goods he bought. Discrepancies in the store inventory will prove fraud. The only way to hide this is to throw out the goods allegedly sold to the customer. Since these goods must now be replaced at market price using fiat instead of NutriCoin—as opposed using an EBT card to make tax-free purchases on the government dime—there would be reduced incentive to engage in fraud and would-be fraudsters would have to invest greater time and energy to develop ways to defraud the system. The government should also conduct snap inspections of NutriCoin merchants to ensure their honesty.

What about customer-to-customer fraud? In this case, a NutriCoin recipient sells his account and PIN to someone else for cash, allowing the buyer to commit NutriCoin fraud. To combat this scenario, the use of NutriCoin could be paired with identification documents, such as a driver’s license, passport or identification card.

As technology advances, NutriCoin wallets may incorporate biometric testing to verify the user’s identity, such as fingerprints or voice samples. After the user keys in his biometric password, the wallet is locked and cannot accept other users or new passwords. The act of keying in a fresh password (but not the password itself) will be recorded on the blockchain, and attempts to change that password without approval will also be noted and recorded. In addition to combating fraud, this also creates increased security for recipients in case their wallets are stolen.


Going beyond restrictions, NutriCoin can nudge people towards healthier living. Welfare recipients are on a tight budget, and so choose the cheapest and longest-lasting foods available. These are inevitably highly-processed and/or junk foods. A steady diet of such so-called food will inevitably lead to obesity, heart disease, diabetes and other illnesses, compounding the recipient’s woes. NutriCoin can prevent this.

The government can pre-designate a whitelist of healthy foods that can be bought with NutriCoin; no other types of food may be purchased with NutriCoin. The smart contract in every consumer-to-merchant transaction will record the type of food the merchant sold, and the merchant-to-government contract will consolidate these transactions for inspection. If there are any discrepancies, the police can again quickly track down the parties involved and examine the merchant’s stock.

While this whitelist should not include high-end luxury foods like organic quinoa or Wagyu beef imported from Japan, it would allow the poor to readily afford fruits and vegetables without worrying about whether they can afford it. By enabling NutriCoin recipients to afford healthy diets, this approach would reduce the incidence of chronic diseases in the long run.

To be clear, the government itself does not set the price for these foods. The merchants are free to decide what price they will set for them and whether to offer discounts for NutriCoin recipients. The government simply ensures that NutriCoin is used for buying affordable, healthy food. More restrictive states will also prevent the use of NutriCoin for buying junk food—which may or may not be desirable, depending on your political preferences.

The beauty of NutriCoin is its adaptability. Once the infrastructure is in place, the basic concepts can quickly be expanded to incorporate other services. This includes public transportation allowances for soldiers, subsidies for utility bills, tokens to pay for higher education or medical fees, NutriCoin POS systems for street vendors and hawkers, subsidised or free meals at food courts and hawker centres, and more. The sky is literally the limit.

There Ain’t No Such Thing As A Free Lunch


The technology described above either exists today or are evolutions of today’s technology. It may even be possible to have a testbed for the system within the next five years. But NutriCoin is not a be-all and end-all solution. It has a few costs which must be overcome or accepted.

The most obvious one is infrastructure. NutriCoin can only exist in a country with a high penetration of smartphones (with an app-based solution), commonplace use of the Internet to deliver services, and a robust information communication technology network. This presently limits its potential to the First World. Fortunately, countries that can contemplate the development of such a comprehensive welfare programme tend to be First World nations themselves.

However, there are still significant implementation costs. Recipients may only need to download an app on a smartphone or carry a hardware wallet. However, merchants will need to integrate this technology into their inventory and point of sales systems, and their bank accounts, requiring new hardware and training. Governments will need to drive development of the blockchain technology, enforce transactions, punish fraud and abuse, assess, approve and monitor recipients, and ensure system uptime. Such costs may well be considerable, especially for small independent mom-and-pop stores in neighbourhoods not served by supermarket chains. While development costs might be reduced through adaptation of existing blockchain and cryptocurrency technologies such as Ethereum, it will not eliminate it.

The next problem is privacy. This blockchain set-up means that the buying patterns of recipients can be monitored, and so can the activities of merchants. Anybody with access to the blockchains and user account databases can quickly and easily identify specific individuals and their patterns of behaviour.

However, since the recipients are beneficiaries of taxpayer money, shouldn’t the taxpayer know how his money will be spent? Perhaps privacy is not such a high price to pay to ensure that NutriCoin is not spent on narcotics or black market guns. One way to safeguard user privacy is to ensure that the records and blockchains are kept and administrated by an independent party, such as a non-profit government-linked organisation or a separate government agency. The police may not access these records without a warrant. Coupled with proper IT security planning, it may enable a reasonable expectation of user security and privacy.

The third issue is cost. Someone has to pay for the programme, and that someone is you, me and every other taxpayer. NutriCoin will be costly. It requires people to manage and administer the blockchain, hardware to keep the system going, police to enforce the law, and a whole new layer of government bureaucracy to ensure proper disbursements of tokens and money and to keep everything running. And that’s not even accounting for the costs of paying merchants when they trade in their NutriCoin.

Furthermore, If NutriCoin were restricted only to healthier foods, the government may have pay out more per recipient than SNAP. Healthy food tends to be more expensive than junk food. This is an especially tricky problem in countries like Singapore, which must import virtually all their foodstuffs from overseas.

Such food costs can be mitigated somewhat with ugly foods. These are irregularly shaped fruits and vegetables that are otherwise perfectly edible. Such foods are usually discarded because they do not meet a merchant’s standards for presentation. They can instead be sold to NutriCoin recipients at a discount, simultaneously reducing costs and food waste. It would not, however, eliminate the issue of cost altogether; it would simply make it less onerous.

A related issue is food deserts. It may be fine in principle to restrict NutriCoin to only healthy foods, but in food deserts, junk food may well be the only kind of food available. This may not be a concern in small countries and city-states like Singapore, but it is a pressing urban issue in America and elsewhere. While the ideal response is to figure out how to push healthy foods into food deserts, this is not a problem NutriCoin alone can solve.

NutriCoin is also not immune to other issues associated with welfare programmes. These include determining eligibility and means testing, how many tokens to dispense to recipients, how the entire programme can be funded and sustained, and so on. It becomes easy—indeed, it may even become necessary—for the government to pass tax hikes to continue funding NutriCoin and other such programmes, which opens a whole new can of worms altogether.

In my opinion, should there be a call for it, Singapore offers the ideal testbed for such a program. Singapore does not have a significant percentage of the poor and homeless people, but it does have a growing number of elderly people who will require assistance. As a First World nation, Singapore has the technology base to implement NutriCoin, and indeed is a regional hub for Ethereum development.

Food deserts are practically unheard of over here, as there is ready access to abundant amounts of nutritious food. Most the population is served not by mom-and-pop grocery stores or corner shops with limited selections, but by major supermarkets—including the Fairprice chain run by the National Trades Union Congress. As the sole trade union in Singapore, NTUC works hand-in-hand with the government. Fairprice enjoys the economies of scale to implement NutriCoin on a national level, its ties with the government would smoothen possible political friction. The major question is whether Singapore can afford to implement this programme, and, of course, whether the government wants to. After all, like with every social welfare programme, the government must pay for everything, and the cost will be passed on to taxpayers.

Blockchain technology offers an opportunity to develop enhanced welfare programmes for the poor and marginalised. The NutriCoin proposal laid out above allows the poor access to healthy foods and merchants to retain their profit margins. However, like all government welfare programmes, NutriCoin carries costs and risks of its own. The question is not simply how blockchains can overhaul food aid programmes and other welfare measures, but also whether such programmes are desirable. Before jumping on the blockchain bandwagon and overhauling welfare programmes, countries would do well to study the technologies and weigh the pros and cons of such initiatives.

Steemit: Liquid Democracy for Social Media

When people think ‘liquid democracy’, they think about politics and decision-making. Steemit, however, offers an opportunity to expand the concept into social media.

A liquid democracy is touted as democracy for the 21st century. Combining the best elements of direct democracy and representative democracy, a liquid democracy allows people to delegate their votes to a proxy who votes on issues for them — but individual voters are also free to withdraw their votes from the proxy if they feel the proxy is voting against their wishes.

This video provides a fuller explanation of the term:


Individual Steemians are free to upvote or flag any post at will. They may create bots to automate the voting process, or delegate their votes to curators. Guilds have their own systems of discussing which posts to upvote or flag. Instead of deciding on policy, though, Steemians collectively decide and reward what they believe is the best content on Steemit.

Lessons for Liquid Democracy

I can vote, you can vote, everybody can vote!

As a model for real-world applications of liquid democracy, Steemit has much going for it. People are rewarded for participating in the platform with tokens that can be exchanged for real currency, incentivising future participation. Instead of outright censoring controversial material, people can simply ignore the post and the writer; the lack of payment tells the writer what the marketplace of ideas really thinks of his content. For abusive content, voters can hide the content through coordinated flagging (‘nuking’). This will not make the content disappear, but flagging does massive damage to the user’s reputation, signalling him to either shape up or get out. And if a rogue users abuse the flagging, other accounts exist to counter these flags.

Steemit is a user-driven platform. Solutions are organically developed from the ground-up. Instead of relying on the mercies of an unaccountable development team, Steemians are free to identify and resolve issues through the platform’s tools where possible. The blockchain prevents content from being arbitrarily removed — at least not without someone noticing — which defeats attempts at government censorship. Unlike other social media platforms like Facebook, Twitter or Reddit, where censorship of dangerous ideas is celebrated, the hand of the dev team or moderators is little-felt at the day-to-day levels.

Social media doesn’t translate cleanly into policymaking. But there are three key lessons from Steemit that apply to liquid democracy.

The first is user empowerment. Steemit is designed around the user, empowering and incentivising participation in the platform. While Steemit’s model allows for weighted votes, in a state or organisation run by liquid democracy every user should only have one vote. This prevents powerful oligarchs from dictating terms to the rest of the nation. However, in a liquid democracy, people are free to delegate that vote to proxies, as in the case of Steemians empowering curation guilds to curate posts on their behalf. This frees people from the need to invest enormous amounts of time and energy into researching topics, and creates opportunities for people with the ability to do so to rise from the crowd. At the same time, voters are also free to vote according to their conscience instead of delegating their vote to a proxy who may not be aligned with their beliefs.

The second is protection of the marketplace of ideas. For a liquid democracy to work, free and fair debate is critical. All ideas must be allowed to participate in the marketplace of ideas, no matter how heinous they may seem to people. Censorship destroys and distorts the marketplace, driving targeted ideas underground but not obviating them. On Steemit, people are free to engage or ignore poor content and conceal abusive ones through flagging; in a liquid democracy, citizens should likewise be free to defeat bad ideas in the cut and thrust of debate, or simply ignore bad ideas into oblivion. The only place for censorship in a liquid democracy is to guard against a greater harm, such as preventing the disclosure of details of a criminal investigation.

The last lesson is decentralisation and division of labour. Nobody can be an expert at everything. In Steemit’s larger curation guilds, different subgroups handle specialised subjects. Steemtrail, for example, has multiple trails dedicated to topics like alternative energy, beer or fiction, operated by curators interested in them. Likewise, in a liquid democracy, political parties may form different subgroups to become the face of the party on topics like the economy, national security and so on. This allows political parties to deploy the best candidates to handle specific issues. While each subgroup would drive the party’s policy with respect to their area of focus, other party members are also free to join in if they are willing and/or qualified to do so. Party members and ordinary voters are free to delegate their votes to these experts. But they are also free to delegate their votes to someone else, or vote directly on the subject. This approach allows everybody’s views to be represented, while creating the environment for subject matter experts to tackle issues they are qualified for.

Lessons for Steemit

Sometimes, you have to be that guy.

Steemit also has a major lesson to learn from liquid democracy: transparency.

In a liquid democracy, proxy voting is supposed to be completely transparent. This ensures voters know how their proxies will vote. The proxy’s voting behaviour is also recorded, letting voters check the proxy’s voting record. Blockchain technology makes this possible by creating a permanent record of a proxy’s voting behaviour and how he intends to vote in the future. Politicians and proxies can count on the media to tell the world where they stand on certain topics and why. Steemit doesn’t offer the same capability.

Users normally trust curation guilds to vote in line with their tastes, but this may not always be so. Using Streemian, users may empower curators to vote on their behalf — up to a point. They may require the curator to vote on certain topics with specific tags, like writing or fiction, or to deny the curator the ability to vote on posts with other tags. This allows users to decide just how much power they wish to give curators. But even this is not sufficient. Here are two examples.

Alice wishes to promote science content on Steemit, so she requires her favourite curator to vote on posts tagged ‘science’. The curator follows her wishes by voting on all posts with that tag — even posts that contain erroneous facts or psuedoscience disguised as proper science.

Barry decides that he will vote on all police-related content himself, so he prevents his curator from voting on posts tagged ‘police’ and ‘crime’. The curator then upvotes and resteems a specific post that makes a false allegation of police brutality. Barry knows for a fact that this allegation is false and would not have upvoted or resteemed it. But as that post is tagged only as ‘writing’ and ‘blog’, Barry’s wishes are not acceded to. He only discovers the curator’s actions only when that post appears on his feed — and wonders what else the curator has voted on without his permission.

Users, curators and writers may do their best, but there will always be gaps not covered by filters. When faced with such scenarios, the easy option is to simply shrug and let the vote stand. Nobody is harmed by the upvote, so why bother? Besides, the user stands to gain curation rewards, and rescinding the vote will reset them.

However, the blockchain makes no distinction between curator and user. A curator may vote on behalf of a user, but on the blockchain the vote is recorded as originating from the user. A person cannot in good faith be expected to be recorded as having voted for a post that goes against everything he stands for, especially since with every vote drains an account’s voting power, reducing the curation reward. Further, should a curator choose to flag a post, there will be negative repercussions for the targeted user. Should a user believe that the target is not guilty, he should be free to cancel his own flag.

In a liquid democracy, every vote belongs to the user, to be given away or taken back as he sees fit. For a direct democracy to work, proxies must be transparent. Their votes and rationale for the votes must be known to all.

As we can see here, Streemian doesn’t do a good job in making votes transparent. The user must first log into Streemian, go to the curator trail and select a specific curator before he can see the voting history. And even that, that history is limited to the last ten votes. This is troublesome for a user, and inadequate.

Going forward, Steemit should introduce measures to enhance curator transparency. At any time, Steemians should be able to quickly access a curator’s voting record with a minimum of clicks, either in Steemit proper or on a third-party platform. The voting record should cover the curator’s history for at least the past twenty-four hours. Curators should justify every vote they make on the public record. When examining the record, users may choose to filter the record by tags for easier reading. Should they discover that a curator has voted against their interests, users are free to cancel their vote at any time. And of course, if a user feels a curator is no long aligned with his interests, he is free to drop the curator and vote manually. Finally, if the platform allows it, users should have the option of recording curator-delegated votes on the blockchain as such (i.e. User delegated Curator to upvote Post X) to distinguish them from manual votes.

Social Media for the 21st Century

Going forward, Steemit is primed to revolutionise social media. It natively encourages users to invest in the platform for the long-term through cryptocurrency. Through the adoption of the blockchain and empowerment of curators, Steemit is now a social media platform that runs along principles of liquid democracy. Steemit’s main stumbling block is lack of curator transparency, underscoring the importance of proxy transparency for policymaking. Should this obstacle be overcome, Steemit could serve as a model for the political evolution of democracies.