It would be spent on drugs or booze anyway, I figured. I'd buy them a hot drink on a cold winter's day or give them my packed lunch, but no cash.
I strongly believe that there's nothing wrong with asking for help if you really need it, but people who constantly rely on others to bail them out will never learn how to help themselves.
I even have a paperweight on my desk that says: "Life is not about how many times you fall down. It's about how many times you get back up."
The kids and I visited Auckland last month, which is where they saw people begging for money on the street for the first time.
It scared them a little at first, but soon they were counting how many they had seen, started reading their signs, and asked me a million and one questions about why people would do such a thing.
Earlier this week, some friends shared an article on their Facebook pages suggesting that the most efficient way to spend money on the homeless might be to simply give it to them.
It was based on a stack of research, and because reading about it opened my eyes, I thought I'd share it with you.
The research findings are not hot off the press but I've only just found out about it.
It started with an example of a research project held on street veterans in London in 2009. The Square Mile in the city, the financial heart of London, has more rough sleepers than any other London area except Westminster.
Some of these people have been living on the streets for more than 40 years. Some take pride in their circumstances and wouldn't want it any other way, and some suffer from drug, drink or mental-health problems, if not all three.
Keeping these people on the streets is costly.
Counting policing, dealing with the justice department, social workers salaries and public health costs, the price tag is estimated to be close to 2.5 million ($5 million) each year.
In 2010, the Joseph Rowntree Foundation and Broadway, London-based not-for-profit organisations that work in the field of poverty and social exclusion, issued a surprising report.
Instead of soup kitchens, shelters and mobile health clinics, they selected 15 homeless people who their field workers had found the hardest to reach, and asked them what it was they needed to change their lives. Each of them received money to purchase exactly what they asked for.
In one drastic bail-out, they each received 3000. It was money in the hand, without conditions. The only question asked was: "What do you think is best for you?"
The wants and needs of the participants were diverse, ranging from a mobile phone to a passport and from a hearing aid to a caravan. None of them wasted the sum on drink or drugs. Most of the homeless men were actually very careful with their spending. After a year, on average, only 800 ($1600) had been spent.
Of the 15 men selected, two refused to take part in the project but of the 13 who did, 11 are now off the streets. They have found shelter, kicked their drink or drugs habit, finished courses, and some have reconnected with family members.
Google and Facebook have also contributed money to Give Directly, a charity which hands out no-strings-attached cash to the poorest people it can find. They provide a wealth transfer service by mobile phone to the extreme poor in Kenya and Uganda. The recipients can use the transfers to pursue their own goals. A similar project in Vietnam in 2006 gave one-off handouts to 550 households; two years later, local poverty rates had fallen by 20 per cent.
These are only small-scale projects, but I think the results are remarkable. It may be a risky approach, but by giving homeless people an opportunity to shape their own future, without interfering and nannying, the savings could outweigh the huge costs of health, police and prison bills.
Even The Economist, a publication focused on international politics, business news and opinion, concluded that the most efficient way to spend money on the homeless is to just give it to them.
I don't think that everyone who is down on his luck should be given a lump sum of taxpayers' money, but the projects I looked into still make sense. I most certainly would support incentives that help people help themselves, as it is likely to save us money in the long run. Give a man a fish, right?