Fighting for freerecyling

40 tonnes per day, only a slight exaggeration?

On CBC Marketplace and elsewhere it is reported that 40 tonnes per day is saved from landfill by freecycling for all TFN groups together.

This particular article adds another titbit to the assumptions made – that each item transferred weighs one pound. With this additional information we have sufficient information to work out whether this 40 tonnes per day is valid or not, or perhaps whether it is an over or underestimate – or whether estimation is at all possible.

Before we go into detail, for those of you who want a shortcut to the answers… To summarise my own initial calculations come out, at best at 20 tonnes assuming all groups are of the size of say Oxford, UK at over 15,000 members. However not all groups are 15,000 strong. If instead we take the average size (arithmetic mean) in say US Central region which is only 750 members as our basis then we only come out at 3 tonnes. That’s 3 tonnes per day is saved from landfill by freecycling for all TFN groups together.

Obviously that is a very different number, around 10x difference.  Basically make a few assumptions and apply these to a reasonable mathematical model and see what number pops out. This is on the basis that one hopes no one will quibble probably because one assumes that no-one is that bothered.

However such matters change when TFN starts taking California State money – public funds – to assist them in meeting their landfill reduction targets.

Now before I continue let me just prefix everything else, I do not know the exact assumptions or formula used, or how TFN’s claims are verified. What follows is my own personal analysis. No doubt by my putting this “out there” some other person may refine or admonish my thinking in some important way. I consider such critique as invaluable, especially in the general context of freecycling supposedly reducing the landfill footprint.

Let us just remind ourselves of the emotional reasons why we freecycle:

  • to reduce landfill,
  • to declutter our own home,
  • for the good of the community,
  • to make way for a new item we want to buy and
  • to help those worse off than ourselves.

Ok, so with freecycling we all sign up to the stated aim of reducing landfill. And in these ecologically sensitive times that adds a good feel-good factor. But in fact a lot of us start freecycling simply because in our eagerness to go shopping we have bought too much, or the wrong things, or even items that don’t last – and that’s what creates landfill. For example, we all know that cars only last for say 5-10, perhaps even 15 years, and manufacturers rely on that – they build in obsolescence so they can sell you a new one. I also remember a local freecycler freecycled their kitchen. How long does the average kitchen last – 10 years? Why not 50 years? I am going to stop with this theme right now because although the above has some truth and that might affect a reduced landfill claim, for the purposes of what follows the assumption is NO EFFECT.

So if we keep it simple we would like to assume that EVERYTHING that is freecycled would actually end up in landfill. Actually there is a problem here also. If we did not freecycle it we might otherwise give it to a local charity, our local church, or directly to needy friends or family. But again let’s, for simplicity, ignore all this. So we assume still this has NO EFFECT.

Now let us begin to think about the “pound” assumption used by Mr Beal. Yes, I obtained a fridge which weighs in at very considerably more than a pound. Also I obtained a table, which similarly was very heavy. But also I obtained some stickers for jam jars, some candles and a range of other smaller items. In summary any person will have their own story to tell about what they freecycled, and it is very very difficult to draw a conclusion as to what the average might be. My gut feeling is that a pound might be a conservative estimate, but if I increase that then where do I stop. Nope, the assumption of one pound weight per item freecycled seems a good place to start, despite it annoying all metrication fans.

And now we can get onto the meat of the model… We next need to decide how we gross up that pound. Well, we could take the number of groups in to account. We could also take the total number of members into account. Finally we could think about the number of posts. The latter is possibly the best place to continue building our model, since at least we know one successful posting results in one item removed from landfill, and therefore following the above assumption we have saved one pound. So we will further assume that the total tonnage is in some way proportional to the number of posts across all groups.

Next we need to think about “posts”. Firstly ever group has posts labelled WANTED, OFFERED and TAKEN. There are some problems here. One is that although every wanted and offered post must initiate a freecycling transaction, some people offer whole garages full of stuff which might make our pound a gross underestimate, but again, let’s ignore that. We do need to make sure though that we do not double-count. If a group has 100 posts in a month then probably a maximum of 50 can be TAKEN posts. Also items are re-offered due to NO SHOWS. Also not everyone posts a TAKEN. Also perhaps only a small percentage of WANTEDs are ever satisfied.

Essentially, without doing a detailed audit, there are lots of issues with trying to decide what proportion of posts could qualify for a valid freecycle exchange. Even a detailed audit of one group may not bear any close similarity to any audit in any other group, especially as we change culture or even time of year.

So let us make a wild assumption. I am going to assume one quarter of all posts in a given month are successful freecycling exchanges. To be clear, that is not saying that 1/4 of anything a given person posts is successful since some people will be very successful, however others may fair badly.

In this we must at all costs avoid double-counting. That is for many WANTED/OFFERs there is at least one TAKEN. When someone offers a list of items it is possible that they post a number of TAKENs. To keep it simple let’s assume that avoiding double-counting is our worst case here. I think that quarter assumption feels near enough.

So we can now easily look at the message history on the home page of each group and copy and paste the number of posts into Excel or other spreadsheet (such as GoogleDocs) and start doing some calculations.

But which year or years, or what months do we take? Obviously when a group starts the number of posts is very low. On the other hand to take a “best” year could vary depend on which group we look at: some groups “mature” and go quiet with less posts, with the owner-moderator team feeling burnt out at a time when they are looking for new blood. My working assumption was to take the ENTIRE message history. This to me looked like a good starting point, since it was quite difficult to think about other choices and every other choice met with more assumptions and questions.

The next is to think about which group, or groups. Well you could theoretically write a computer program and visit each and every group. Or you could just take an “average” group. I took, for my average group Brighton, CO. Let me explain how I arrived there. This was in fact very close to the numerical average (arithmetic mean) of all groups in the US Central region at around 800 members. Yes there are some groups over 10,000 but there are also a lot of groups below 1,000, and a lot of those below 800. The arithmetic mean came out at 750 and the name Brighton appealed to me. The appeal is because I am actually going to Brighton, UK for a couple of days next weekend. So there you go, I assume Brighton, CO is the average freecycle group – worldwide. Yes, did you notice I grossed up the US Central into the worldwide “norm”, so that’s another assumption.

Finally in going worldwide we need to know how many groups there are worldwide. Fortunately that’s reasonably easy – let’s assume 4,000.

So to summarise how the calculation is done, here is my take:

  1. Capture the number of posts the “average” group over its entire history from the home page
  2. total those
  3. multiple by the total number of groups (4,000)
  4. divide by the number of years the group has been running
  5. divide by the number of days in the year
  6. divide by 4 to ensure we avoid double-counting offer+take as well as failed offers and repeat offers etc
  7. convert number of items freecycled to number of pounds – 1 item freecycled =1 pound saved from landfill
  8. convert pounds to tonnes, 1 pound = 0.00045359237 tonne

And there you have it. The entire model.

Please pull this apart and/or find out where my arithmetic is wrong. What assumptions would you make?

My bottom line is, it is a fun mathematical game to play, but there are a lot of assumptions and any answer is likely to be out by a mile, or 2,000.

If you know of any research that can extend or validate or invalidate the above then please do add grist to the mill.


October 14, 2007 - Posted by | freecycle

No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: