In an interesting contrast to yesterday’s piece about why Standard Motor Products keeps much of its manufacturing in America, the New York Times has a lengthy article on why Apple products are not made in the USA:

One former executive described how the company relied upon a Chinese factory to revamp iPhone manufacturing just weeks before the device was due on shelves. Apple had redesigned the iPhone’s screen at the last minute, forcing an assembly line overhaul. New screens began arriving at the plant near midnight.

A foreman immediately roused 8,000 workers inside the company’s dormitories, according to the executive. Each employee was given a biscuit and a cup of tea, guided to a workstation and within half an hour started a 12-hour shift fitting glass screens into beveled frames. Within 96 hours, the plant was producing over 10,000 iPhones a day.

Both Apple’s products and some of Standard’s products require skilled labour and fancy machines to produce. But to me it seems the difference is that Standard’s products are motor parts that have a relatively predictable demand and are not R&D intensive, so a high degree of manufacturing flexibility is not required. In contrast, the need to scale up production rapidly to meet demand for Apple products, and accommodate last-minute design changes, means Apple needs flexibility most of all.

Adam Davidson has a fascinating article in The Atlantic about manufacturing in America, and indeed why this continues to exist in spite of low cost competition from China and elsewhere. The article focuses on Standard Motor Products:

Standard will not drop a line in the U.S. and begin outsourcing it to China for a few pennies in savings. “I need to save a lot to go to China,” says Ed Harris, who is in charge of identifying new manufacturing sources in Asia. “There’s a lot of hassle: shipping costs, time, Chinese companies aren’t as reliable. We need to save at least 40 percent off the U.S. price. I’m not going to China to save 10 percent.” Yet often, the savings are more than enough to offset the hassles and expense of working with Chinese factories. Some parts—especially relatively simple ones that Standard needs in bulk—can cost 80 percent less to make in China.

Via Longform.org

Another interesting thing about iBooks – the maximum price is 15 US dollars.

The maximum price of an iOS app is 999.99 US dollars.

Apple’s new iBook Author software looks pretty cool. If I was going to write a book, I’d certainly be tempted to use it.

It’s interesting that the license agreement contains a little fuck you Amazon … all books created with iBook Author must be sold via the iBookstore (if you give the book away for free you can do whatever you want).

As far as I know, Kindle has a much bigger share of the ebook market than iBooks. But iPad has a much bigger share of the tablet device market. It’s typical and expected of Apple to try to leverage its share in devices across to content by providing cool tools to content developers. And cut out the publishers in the meantime – very disruptive.

Seems like a good response would be for Amazon to release similar software for authors. Not sure if they can match Apple’s polish though.

Update: Josh Gans points out the author agreement is not quite exclusive, you can take the content produced by iBooks and assemble it in another program and sell it through another channel.

The supermarket in my office building has two queues. One feeds a bank of six human checkout operators and the other feeds a bank of eight self-checkout machines.

In my experience so far, the human operator queue moves significantly faster than the self-checkout queue. Furthermore the self-checkout queue is noticeably longer than the human operator queue. This is despite there being two more self-checkout machines than human operators.

It seems to me that people systematically over-estimate their own ability to efficiently self-checkout. Leave it to the pros, I say!

Continuing the theme … how to give me customised discounts based on my purchasing patterns? One way is to send me vouchers in the post, but that’s unsexy and expensive. And I’ll forget to bring the vouchers or lose them. Or I might give them to someone else, defeating the purpose of the targeted discount.

Let’s see if we can use an app to solve these problems … Vouchers can be pushed to the phone, but then how to validate them at the checkout? I suppose I could get a code to give to the cashier, but that’s clumsy, and again I might give the code to someone else. We need to make arbitrage a bit more difficult. We could do that by pushing the vouchers to my phone only when I enter the supermarket, but how to know this has occurred? GPS isn’t that accurate …

So how about this: When I enter the supermarket, I scan a QR code on my phone that is posted at the entrance. This code changes daily, and prompts the app on my phone to download my targeted offers for that day and show them to me. This also alerts the supermarket’s checkout computer that I’m in the store, and activates the discounts when I swipe my loyalty card at the checkout.

I still think this is a bit clumsy, what we really need is a better GPS system …

The supermarket I regularly use has a loyalty card which allows them to track my spending patterns. But they don’t seem to do anything terribly useful with all the data, they just simply send me some ‘reward’ vouchers from time to time. That will make me a little bit more sticky, but surely they can do better …

A smart thing to do would be to figure out products that I don’t regularly buy and give me special discounts on those. I obviously have high demand for the products that I do regularly buy, so there’s no point to give me discounts on those, my demand is inelastic. But give me a discount for something I have low, elastic demand for, and I might buy it. If chosen carefully, this could increase my total spending (rather than just cannibalising my existing spending) and increase revenue for the supermarket.

A large supermarket opened in the ground floor of my office building, giving me the option to buy ingredients to make lunch at a significant discount from what I would usually pay to buy lunch in the city.

Of course this raises the question of why I didn’t do this already when I visited the local supermarket near to my home. You might say I’m lazy, but the economist in me has other answers -

I only visit my local supermarket infrequently, so I would need to buy larger quantities, with the accompanying risks and costs of spoilage and storage, plus less daily variety in my lunch.

It’s a little inconvenient to bring the lunch with me on the bus, and I might forget.

I have to decide what I want for lunch in advance and give up the option value of deciding at the last minute.

So, apparently I was willing to pay a price premium because of these costs. But now having a supermarket very near to my office means I can have the best of both worlds. The supermarket should be able to charge a higher price and extract some of the surplus, but they don’t seem to be doing that, so it’s all gain for me!

A tweet today got me thinking a little about mobile number portability (MNP) … The conventional wisdom is that MNP increases competition in mobile markets as it allows customers to switch more easily between networks. However there is also a limited supply of ‘good’ numbers on any given network (123-4567, 888-8888, etc). A new network with its own number prefix is able to replenish the supply of ‘good’ numbers and some customers who have a ‘bad’ number on their old network will be willing to switch even if they have to change their number.

So MNP is not necessary for the new network to compete for all customers, and in fact the new network is at an advantage in terms of attracting customers with ‘bad’ existing numbers. (Whether or not customers with ‘bad’ numbers are valuable customers is another question, however.)

On top of this, unless there’s some system for alerting customers about which network the people they are calling belong to, MNP possibly introduces some confusion about the cost of making a call, if there are different prices for calling on the same network versus calling other networks. This could have effects on customer behaviour and competition between networks that are more subtle … without a formal economic model it’s hard to be sure.

Anyway, I don’t have any strong conclusions here, just that there are some aspects MNP that might not be immediately obvious. And it’s interesting how a simple tweet can get you thinking about all sorts of things.

I found these targets published by a government organisation that shall remain nameless:

• 80% of customers served within 10 minutes
• 80% of telephone calls answered within 30 seconds
• 80% of email inquiries responded to within 7 days

Why is it considered acceptable to take SEVEN DAYS to answer email, while walk-in customers should not be kept waiting for more than 10 minutes and telephone callers not more than 30 seconds?! If anything it should be the other way around …