Similar to a mountain, this represents the transaction amount and price; these two multiplied are the transaction volume, all in one mountain. Now imagine the transaction cost as the water around this mountain. What we suppose here is that the transaction cost is above this line between the blue and green, that is the light blue part is not submerged in the water. What can be transacted is the part above the water. Now that the transaction cost is down, as we imagine, the water is down here between the deep blue and light blue. Apparently, what we thought worth buying, originally the green and red area got extended including now the light blue part. What we need is the three-dimensional model, and I’ll do the evaluation with this model.

Therefore, when we talk about the “long tail”, we should also imagine it in the three-dimensional space, instead of the two-dimensional space. Now we see, as transaction cost is reduced, more transactions can be done, much larger than we thought.

We need to evaluate the transaction cost, though, heading for more technical details that are of importance. We’ve got some figures. First, the transaction cost of industries which can be obtained from the accounting materials and statistics of those enterprises. The figure is around 1% of the retail price. Another figure is the transaction cost of commercial, close to the commercial added value, which we also have. Here’s a nationwide input-output table of 2010, where the commercial added value is about 59%, which is 59% of the retail price.

Second, we need to consider the consumers’ transaction cost. As I just mentioned, both the industrial and commercial transaction costs are included in the price of products, in the form of currency. As we buy those products, we pay for the transaction costs with money. However, the transaction cost of consumers does not take the form of currency, or entirely of currency. It is in the form of the time consumers spend on their way to the store, and some other difficulties that they have to deal with, which can only be estimated according to our experiences. Of course, what I have today is my own judgement. As products vary, the transaction cost of consumers varies. I’d like to take books as an example, since it’s a standardised product that I purchase a lot. I bet many of us do, too. As I estimate, the transaction cost of a book is its retail price, that is 100%. As we have to evaluate the transaction cost of other products too, I’ll tune it down to 50%. When we add up the three types of transaction cost, it accounts 110% of the retail price of the product, which is high. [Page]

This is just a rough estimation that is open to question. I’m just providing a way to illustrate this issue here.

Now that we have online bookstores and booming e-commerce, what happens to the transaction cost, then? Let’s say that not much has changed to the industrial transaction cost, which is just 1%. What’s more important are the other two transaction costs, one of which is the commercial transaction cost. We could see it as a change to the commercial added value rate. As I just mentioned, we’ll take a book as a standardised product for the simplicity of the analysis. In the past, I go to bookstores to buy books according to the price tags. I’d pay 20 RMB for a book as the book is priced that much. But now, I purchase books from Dangdang, an online bookstore. Normally there is a discount of 25% off, but of course Dangdang gets those books at an approximately stable discount rate from the publishing houses. A 25% discount is 25% off the commercial added value. Therefore, the commercial transaction cost is down by 25% from 59%, making 34%. That is how much the commercial transaction cost is lowered.

As for the transaction cost of consumers, it is rather difficult to estimate. I’ll estimate according to my experience, and I encourage you to make your own estimation. In the past, I usually went to Wansheng Book Market. It normally took two hours. As I had to roam through the book market for another two hours, plus there were traffic expenses, the transaction cost for me was high. Now that I have my iPad and I can buy books online, the transaction cost is one tenth of the cost before. These are estimated figures. If the previous transaction cost was 50% of the retail price, now it’s only 5%.

As I estimate, if we adopt the two-dimensional model, an increase of market volume could reach 80% thanks to online shopping, which is 80% of sales volume. However, if we use the three-dimensional model, the mountain model as above, this increased sales volume equals two previous markets. That’s just rough estimation, but the gist is there. And the sales revenue could increase as much as that of one and a half markets.

That is to say, if there’s sales revenue of 30 trillion RMB, with the online shopping at disposal, it could reach 75 trillion RMB.

How did this change take place? As I see it, people begin to seek for alternatives in the beginning, such as new ways to buy books. Apparently buying books online is more convenient than going to a real bookstore. There are many other products that I don’t bother buying in a real shop now. In the past, I’d rather skip it if it was too much trouble to go to the shops. For instance, if I want to buy a drone and don’t know where to buy it, all I need to do is just click the mouse and buy it online. This is the new increased demand.

This process can also be roughly estimated in this chart: the blue area is the alternative demand, and the red area is the new increased demand. These are the real figures over the time period of 2008 to 2015. As we can see, the new increased area is getting bigger and more important, [Page] whereas the alternative area is getting smaller. This is a structural change.

What’s our estimation of the online retail then? We estimate that it has increased by an average about 72% on a yearly basis from 2005 to 2015. But it slowed downed a little to 38% in 2015. Therefore, I wouldn’t be too optimistic for the future. There is not likely to be rapid growth in online retail in the future, and it’s slowly growing.

So, a cautious estimation is that the online retail will keep growing at 25% annually from now on. The straight line in the chart is the estimated figure, and the figures before were all real historic data, the highest was a growth rate of 120% in 2008.

The online retail revenue is growing fast and weighing more in the total retail revenue. In this chart, the grey area is the non-online retail revenue, that is conventional retail revenue; and the blue area here is the online retail revenue. As we can see, the blue area has been increasing over the years. This is my estimation for 2025, when online retail is likely to constitute 40% of total retail revenue.

By 2025, online retail revenue is likely to be 1.5 times that of 2014, and the new online retail volume is likely to amount to 1.5 times of the total market volume.

Be advised, this doesn’t mean that we’ll be there in a blink of an eye. It is a process that takes time. As we all know, online users and consumers are growing in number. It’s not that as we have Taobao, an online shopping platform, everyone starts using it instantly. 

At last, I’d like to provide you a rough estimation figure. Do not take it as a true fact, but rather as a general picture of how the landscape will change. In this chart, the orange line here is the boost to GDP from the increase of online retail. Thanks to the increase of online retail, our GDP growth increases by a few percentage points.

Just to clarify, there are two aspects involved, one is the commercial added value as I mentioned, and it is mainly the online shopping commercial added value. However, it differs from the 59% of added value rate increase earlier. The reason is the reduced transaction cost. As the retail price is also decreased, the 59% and 34% added value rate do not apply because the new price is not taken into account. My estimation is about 46%, that is the added value rate.

Another aspect is the industrial added value of industrial products that needs to be taken into consideration.

This is just a static estimation. It’s difficult to give a dynamic estimation, because online shopping has brought about tremendous change to the way of production, division of labour, and technological innovation. In the static estimation, the New Economy contributed at least 1% to the GDP growth in 2015.

I think this figure could increase to 3.3% by 2025.

As we saw in the charts earlier, the GDP growth rate was pretty high in the past few years, and the turning point was 2015. After that, it goes up again thanks to the New Economy that features mainly the growth of online retail. This is the general picture. [Page]

（This article presents remarks made by Professor SHENG Hong at the Seminar on “An Estimation of the New Economy’s Contribution to Economic Growth” held at Unirule Beijing Office on September 27th, 2016.）