Wednesday, November 20, 2013

(Extremely) conservative economic analysis of bakfiets ownership

In the first twelve months of owning a bakfiets I rode a little over 4,000km, averaging about 80km/week.

So I began wondering about quantifying the cost-effectiveness of the bike.  Was I saving money, or was it a false economy?

Short answer

Yes, I am almost certainly saving money at 80km/week.

Averaged over five years, the petrol savings alone are enough to break even.

If you include the variable depreciation component as modelled by the VTPI, then it becomes clear that the break-even point, based on fuel and variable depreciation alone, is at between XXX and XXX kilometres per week.

If, like ourselves, you can avoid owning a second car then the savings are obviously much bigger, as a bakfiets costs about $1,000 per annum including the cost of an expensive personal loan to buy the bakfiets, and making the further extremely conservative assumption that it is worth nothing after five years.

Long answer

Let's make a few purposefully conservative assumptions.  The idea is if these figures work out positive for you, then you should be able to fairly safely assume that owning a bakfiets will save your family money.

1. You already own a car, that consumes approximately 10L/100km.
2. You will retain the car, and will still service it as often as now.
3. You will displace some kilometres from the car to the bakfiets.
4. The sunk-capital costs of the car will not drop.
5. Petrol costs about AU$1.30 - AU$1.60 a litre in Adelaide now.  This equates to AU$0.13 - AU$0.16 per kilometre.
6. Including all accessories, delivery and assembly, a good Butch bakfiets costs about AU$3,500.
7. Interest cost on a bakfiets will be between 4.5% (lowest rate home mortgage) to 10% (expensive personal loan).
8. A bakfiets will cost about AU$120 a year for servicing and incidental items (replacement tyres etc).
9. Dutch bakfiets have a design-life of about 20 years, but let's be really conservative and say that you we expect only five years of useful life, after which it has the implausibly low residual value of AU$0.

This is a very conservative set of assumptions, as eradicating short trips will increase the life of most vehicles through reduced cold-operating wear, as well as pushing out servicing intervals.  But again, we want to be exceptionally conservative here.

In other words, we are looking at the pure incremental cost and savings of owning a bakfiets.

Petrol, as mentioned costs between AU$0.13 to AU$0.16 per kilometre at current prices.

The above assumptions require a five-year loan for AU$3,500 with a prevailing interest rate of between 4.5% and 10% per annum.  That loan will cost between AU$15.04 and AU$17.12 per week. The annual servicing costs of AU$120 will add another AU$2.31 per week.  Thus the total weekly cost of buying and operating a bakfiets over five years will be between AU$17.35 and AU$19.43 per week.

The above implies that you will need to displace between 109 and 150 kilometres per week to be practically certain that you will come out financially ahead compared by owning a bakfiets.

This is more than the 80km/week that I ride to work and doing the morning school/child-care run.

Let's now consider this over the coming five years, assuming that inflation remains at about 3%, and petrol prices will, on average, at least match inflation compared with current prices (which seems reasonable based on historical data).

YearExpected petrol price per kmWeekly kilometers to break even
113c - 16c109 - 150 / week
213.39c - 16.48c106 - 146 / week
313.79c - 16.97c103 - 142 / week
414.21c - 17.48c100 - 137 / week
514.63c - 18.01c97 - 133 / week
615.07c - 18.55c13 - 15 / week *
* With no loan repayments from the 6th year onwards, there is only the operating cost of the bakfiets to consider, hence the dramatic drop in operating costs.

Again, this analysis is quite conservative, as the cost of petrol typically trends up at a rate nearer to 10% per annum, but with some downward corrections from time to time when the global economy tanks.

Taking historical petrol price trends into account

Looking at the best data set I could find, which records the price of fuel in the USA since 1990 (see here), we see that the cost of fuel increased from US$1.191 to US$3.262 per US-gallon between August 1990 and November 2013, for an average annualised increase of 4.5%, compared with US inflation which averaged 2.7% during the same period (see here).

So let's calculate the outlook where we assume that petrol will increase at 4.5% per year instead of 3% per year:

YearExpected petrol price per kmWeekly kilometers to break even
113c - 16c109 - 150 / week
213.59c - 16.72c104 - 144 / week
314.20c - 17.47c100 - 137 / week
414.88c - 18.26c96 - 131 / week
515.50c - 19.08c91 - 126 / week
616.20c - 19.94c12 - 14 / week *
* With no loan repayments from the 6th year onwards, there is only the operating cost of the bakfiets to consider, hence the dramatic drop in operating costs.

These figures are still on the high-side of realistic, given that I only average 80km/week. But as I have made clear above, the figures are purposefully very conservative.

Taking realistic residual value of the bakfiets into account

For example, it is likely that a bakfiets would have a residual value of around AU$1,000 - AU$2,000 after five years, and assuming it was well enough maintained to still work, AU$500 would be the lowest realistic figure here.  Recycling the capital at the end of the five years (lets assume AU$1,000 resale value for the bike at the end) to payout the balance of the loan would reduce the weekly loan repayment to between AU$11.61 and AU$14.16 per week.  Adding the AU$2.31/week operating costs for the bakfiets back in, the total weekly cost would be AU$13.92 - AU$16.47, .

So let's see how that works out:

YearExpected petrol price per kmWeekly kilometers to break even
113c - 16c87 - 127 / week
213.59c - 16.72c83 - 122 / week
314.20c - 17.47c80 - 116 / week
414.88c - 18.26c77 - 111 / week
515.50c - 19.08c73 - 107 / week
616.20c - 19.94c12 - 14 / week *
* With no loan repayments from the 6th year onwards, there is only the operating cost of the bakfiets to consider, hence the dramatic drop in operating costs.

That one simple change of assuming an AU$1,000 residual value at the end of the five years reduces the kilometres to about what I do now -- still without relying on any of the "fixed" costs of owning a car reducing.

Taking variable car depreciation costs into account

The VPTI suggest that the variable depreciation cost per mile of travel is about US$0.055.  That is, driving one extra mile will, on average, reduce the resale value of a five year old car by 5.5c.  On newer cars the effect is even greater.  Thus we can add 5.5c/mile = 3.44c/km to the fuel cost and recalculate the impact:

YearExpected petrol+depreciation cost per kmWeekly kilometers to break even
116.44c - 19.44c72 - 100 / week
217.03c - 20.16c69 - 97 / week
317.64c - 20.91c67 - 93 / week
418.32c - 21.70c64 - 90 / week
518.94c - 22.52c62 - 87 / week
619.64c - 23.38c10 - 12 / week *
* With no loan repayments from the 6th year onwards, there is only the operating cost of the bakfiets to consider, hence the dramatic drop in operating costs.

Given that the worst case cost of owning a bakfiets (on a 10% loan and no residual value at the end) is only $1,014 per year, you don't need to save much on parking costs, less frequent car servicing and other incidentals to reduce the break-even point.  And that doesn't include using the bakfiets to avoid trailer hire, taxi fares to the airport, let alone being able to get rid of that second car (or in our case, not needing one in the first place).  We have also completely ignored the social, emotional, family and other non-economic benefits of bakfiets ownership.

In fact, my observation is that riding around Adelaide you can collect an average of one 10c deposit container every two kilometres, thus offsetting the cost of running by 5c per kilometre.  5c x 4,000 kilometres = $200 per year -- sufficient to easily cover the non-capital operating costs of the bakfiets.

Saturday, November 9, 2013

Cargo bikes at the Adelaide Christmas Pageant

This is not the first year that we have ridden in to the pageant with the kids.  This year it was especially appealing, as our state government is still in the process of upgrading the local railway line, and driving is never a good option to an event that attracts about 20% of the total population of the state in any given year.

This year the kids used the seats in the box to increase their elevation so that they could get a better view:

Caleb's attention span was a bit short, so he and I walked around a bit, allowing Mummy to climb in with Isabel who together had a lovely "girls time" watching the floats:

The Adelaide Christmas Pageant is apparently the largest in the world (179 floats, bands and other features this year, apparently), and is certainly lots of fun for little people (and their accompanying big people).  I didn't get many shots of the floats, but

Walking around with Caleb, I had a good chance to see what other bikes people were riding there.  It was great to see a lot of people had ridden in on various "normal" bikes, but I was pleasantly surprised to see this year was the number and variety of cargo bikes that were there.

I don't remember seeing any other than ours last year, but this year there was quite a variety.  Here are pictures of the ones I found within a few hundred metres of our spot on South Terrace where the pageant starts.  I have no idea if there were more (or how many more) along the couple of kilometres of the remainder of the route.

Another cargo long, just like ours, but with reflector strips on the side.

One of several trikes.

Caleb inspecting another trike with fun space-invaders decoration.

Some other kids loaded in their trike. I don't remember if they were coming or going.
Yet another trike.
The number of trikes was intriguing.  Some of them were from families with 3+ children, which makes sense for the space.  But at least one parent I spoke to, on learning I had a two-wheel cargo bike, expressed what I took as some regret that they hadn't bought a two-wheeler, as he had realised that the trikes are not as stable at speed.

I remember when I was contemplating getting a cargobike that I was leaning towards a trike, and it really was just someone who made me realise that a two-wheeler is the better option (unless you have too many kids to fit).  They simply asked me two questions: 1. Can you ride a bike? and 2. If so, then why are you thinking of getting a trike?  The implication being that the two-wheelers are a better option, unless there is some reason why you can't ride or use one.

I am inclined to support their view.  Two-wheelers are no wider than a normal bike, substantially lighter (possibly the only sentence where a cargo long and "lighter" can be used together ;), handle well at speed (we have had ours up to 65km/hour with kids on board, admittedly down a big hill. They are also typically a bit cheaper.

Not strictly cargo bikes, but I still approve of this method of transportation, and unlike us, they were allowed to join in the pageant.

There were a few bike+trailer combinations, but there were more cargo-bikes than trailers.

There was even a bullit, fitted out to seat one fast passenger.  Bullit's are much faster than most other cargo bikes, but with more traditional bike parts (which is part of their appeal for many) are more effort to maintain, like a "normal" bike.
 We also managed a few shots of some of the 179 floats and bands as we walked about.


There was even a float with a bicycle theme:

And of course Father Christmas and his reindeer are the cavalry, following the floats to the "Magic Cave" on North Terrace:

Then it was time to get ready to ride home: