How to "Run" an NGO in Hong Kong

I want to say upfront that I am writing this from inside the sector, not from outside it. I am not writing about a competitor. I am writing about a set of habits that I have watched become normal, including habits I have had to check in myself.

First, understand that nobody is checking

Before the playbook, you need to understand the terrain, because the whole thing rests on it.

Hong Kong has no charity regulator. There is no single body that registers charities, monitors them or has the power to look into how they actually operate. The Law Reform Commission recommended setting up a charity commission after a six year review. The government declined, on the basis that there was no consensus in the community for a comprehensive framework, and introduced some administrative measures on fundraising transparency instead.

There is also no statutory definition of what a charity even is. Under the Inland Revenue Ordinance the terms still rest on English case law from 1891. Poverty relief, education, religion and a fourth catch all category. That is the test.

What exists instead is a scatter of oversight. The IRD grants tax exemption under section 88 and keeps the closest thing we have to a list. The Companies Registry handles you if you are a company limited by guarantee. The Societies Ordinance handles you if you are a society. The Social Welfare Department has views if you are subvented. The ICAC publishes a best practice checklist.

Five different bodies, each asking a different question, none of them asking the one that matters.

Nobody asks whether the two hundred young people in your annual report are two hundred young people.

Now the playbook.

Rule one: buy the room with a scholarship

Scholarships are the cheapest reputation you will ever purchase.

You hand over a sum of money in a hall with a banner behind you. Someone photographs it. The photograph does more work for you in one afternoon than three years of casework ever will, because giving money to a bright student is legible to a donor in a way that structural disadvantage is not.

You have now bought two things. Public standing, and a cohort.

Rule two: recycle the cohort

Here is where it compounds.

Your scholarship recipients are now available to you. Invite them to your leadership programme, the one funded by a corporate foundation. Invite them to your mentorship scheme, the one funded by government. Invite them to your community day.

Each programme reports its participant numbers separately. Each funder sees a healthy figure. No funder can see the other funder's list, and no regulator is positioned to compare them, so nobody will ever discover that it is the same seventeen faces cycling through four line items.

You are not lying. You are reporting contacts and letting everybody read them as reach.

I want to be fair here, because there is a real version of this. Working with the same young people over several years is often better practice than reaching new ones every quarter. Depth beats spread. The dishonesty is not in the repeat participation. It is in reporting repeat participation as new reach and never saying which one you mean.

Rule three: let them feel they owe you

This is the part that I find hardest to write about, because it is the part that works on good people.

Give a student a large sum of money and something happens that no contract could achieve. They want to give back. So they volunteer. They show up at your events, they staff your registration desk, they speak on your panels, they appear in your recruitment video.

And you count those hours. You count them as volunteer contribution, as social value, as community engagement.

Look at what has happened to the money. It left as charitable expenditure. It came back as unpaid labour, and you booked it a second time as impact.

The student is not being cynical. That is exactly why it works. They are being decent, and their decency has been converted into an asset on your reporting. Somewhere in there volunteering stops being something a person chose and becomes something a person owed, and then eventually it becomes a box they tick because they were told it was expected.

Ask yourself honestly whether a person who has just received a large sum from you is in a position to say no to you. Then ask whether you have ever set up a situation where they could.

Rule four: hold the data

Run the programme, give out the freebies and collect everything. Names, schools, household income, languages spoken at home, what they want to be, what they are struggling with.

Then keep it.

Do not share it with the school. Do not share it with the other organisation working two streets away with the same families. Do not publish anything except the parts that photograph well.

The data is not for the community. The data is for the deck. Held tightly enough, it becomes the thing that makes you the authority on a population you have surveyed but not served.

And when someone proposes a shared referral system across organisations, be supportive in the meeting and slow afterwards.

Rule five: pay little, brand everything

When you partner, negotiate down. Offer the smallest possible contribution.

Then make the logo non negotiable. Front of the banner. Top of the press release. First name in the caption.

The partner did the work. You did the visibility. Visibility is what gets funded next year, so you have made the better trade.

Rule six: consult, then own it

Hold a closed door session with your service users. Call it co-creation.

Let them tell you what is actually wrong, because they know, they have always known, and nobody has ever asked them properly.

Take the notes. Do not pay them. Do not credit them. Six weeks later, present their analysis as your insight in a funding proposal.

There is a name for this now. Participation washing. The room existed so that the report could say the room existed.

Rule seven: keep the worst story on rotation

You will have one family whose situation is genuinely dire. Tell that story everywhere.

Never mention that it sits at the extreme end of what you see. The audience will assume it is typical, and you will not correct them, because a typical story does not move a room.

Over time you will have taught your donors, your partners and the public that a whole community is defined by its worst month. You will have made your funding case by making the people you serve smaller.

Rule eight: stay downstream

Do not go near the root cause. Root causes involve policy, they involve years, they involve saying uncomfortable things to people who fund you, and they do not produce a photograph in March.

Stay with the symptom. The symptom renews annually.

Put the money into marketing instead. Marketing has a measurable return. Structural change does not, at least not on the timeline your annual report runs on.

Rule nine: let funding do the vouching

Collect enough government contracts and corporate partnerships and something quietly shifts. People stop asking whether you are good and start assuming it, because surely all those funders checked.

They did not check. They saw the other funders.

That is the whole trick. In the absence of a regulator, funding volume becomes the proxy for credibility, and credibility attracts funding volume. You are now the organisation people go to, and you got there without anyone ever examining an outcome.

Now the honest part

Every rule above describes an organisation that is accountable upwards and not downwards.

Upwards means to funders, boards, government and the press. Downwards means to the people whose names are in your database.

The sector is not short of integrity. Most people I have met in it are doing hard work for modest money because they care. What the sector is short of is any structure that rewards them for it. When the only party who can end your organisation is the one writing the cheque, you will optimise for the cheque. Not because you are corrupt. Because that is what the incentive is.

So it will not be fixed by better intentions. It will be fixed by a few boring things.

Report unique individuals served, not contacts and state your method. Publish who funds you and how much. Pay the people whose insight you build programmes on. Share anonymised data with other organisations serving the same families and do it before you are asked. Say what your programme did not achieve. Do not use a person's worst week as a fundraising asset. And when a young person you gave money to offers to volunteer, make sure they can walk away from you without cost and check whether they know that.

None of this needs legislation. All of it can start tomorrow.

The regulatory vacuum is real and it should be closed. But I do not want to wait for a charity commission to tell me not to count the same student four times.