The shift of trade powers
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Subtitle |
| |
| Title of Series | ||
| Number of Parts | 295 | |
| Author | ||
| Contributors | ||
| License | CC Attribution 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/43569 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
FOSS4G Bucharest 201915 / 295
15
20
28
32
37
38
39
40
41
42
43
44
46
48
52
54
57
69
72
75
83
85
87
88
101
103
105
106
108
111
114
119
122
123
126
129
130
131
132
137
139
140
141
142
143
144
147
148
149
155
157
159
163
166
170
171
179
189
191
192
193
194
195
196
197
202
207
212
213
214
215
216
231
235
251
252
263
287
00:00
Shift operatorNatural numberTable (information)File formatData analysisExploratory data analysisCodierung <Programmierung>Physical systemCoding theorySelf-organizationProduct (business)Standard deviationNumerical digitData miningPhysical systemTraffic reportingQuicksortStandard deviationDigitizingCodeDataflowDescriptive statisticsComputer animation
01:14
Self-organizationNumerical digitPhysical systemStandard deviationProduct (business)Table (information)File formatExploratory data analysisData analysisCodierung <Programmierung>Coding theoryVisualization (computer graphics)SoftwareDatabaseData analysisModule (mathematics)Run-time systemBitExtension (kinesiology)Open sourceTable (information)SoftwareKey (cryptography)Right angleWebsiteExploratory data analysisComputer animation
02:21
Data analysisLine (geometry)Endliche ModelltheorieData structureBitWorkstation <Musikinstrument>Visualization (computer graphics)LaptopLibrary (computing)Module (mathematics)Lecture/Conference
03:05
Integrated development environmentSoftwareVolumeOnline chatLine (geometry)NumberVolume (thermodynamics)Term (mathematics)Library (computing)Parameter (computer programming)Traffic reportingPoint (geometry)Computer animation
04:02
Doubling the cubeMathematicsNetwork topologyLecture/Conference
04:28
SoftwareIntegrated development environmentVolumeInversion (music)Online chatSource codeNumberElectric currentDatabaseFrame problemCASE <Informatik>Power (physics)Library (computing)State of matterParameter (computer programming)Term (mathematics)RankingLaptopProcedural programmingMiniDiscData storage deviceSet (mathematics)Standard deviationVisualization (computer graphics)Computer animation
06:47
NumberProcess (computing)Digitizing
07:14
State of matterSource codeNumberNumberDrop (liquid)QuicksortShift operatorPower (physics)DigitizingBitWordNatural numberComputer animationDiagram
07:59
Source codeNumberScalable Coherent InterfaceSummierbarkeitAverageState of matterNatural numberElectronic mailing listNumberData mining1 (number)Multiplication signPosition operatorMusical ensembleClosed setSoftware testingCue sportsRight angleWordDigital photographyDirected graphLecture/ConferenceComputer animationDiagram
10:46
NumberAverageConcentric1 (number)Subject indexingCASE <Informatik>Shared memoryNumberAverageMixed realityObservational studyScatteringCartesian coordinate systemIntegrated development environmentPrice indexPlotterRight angleArithmetic meanRing (mathematics)Computer animation
13:48
MathematicsSubject indexingComputer animation
14:29
VolumeSubject indexingProcess (computing)MereologyShift operatorComputer animation
14:53
Table (information)MathematicsBit rateMereologyProcess (computing)Insertion lossSet (mathematics)Arithmetic meanNatural numberBit rateService (economics)AreaLevel (video gaming)Disk read-and-write headSoftware developerFreewareSinc functionComputer animationLecture/Conference
16:30
Shift operatorCalculationMathematical analysisGravitationMassRaster graphicsFunction (mathematics)Physical systemLibrary (computing)Arithmetic meanService (economics)Shift operatorRaster graphicsSet (mathematics)Population densityGravitationMassPower (physics)Cartesian coordinate systemPoint (geometry)Different (Kate Ryan album)MereologyGradient descentLecture/ConferenceComputer animation
18:22
CodeSlide ruleDifferent (Kate Ryan album)Lecture/Conference
18:51
Function (mathematics)CalculationSummierbarkeitScripting languageGravitationLoop (music)Raster graphicsPoint (geometry)Table (information)PolygonPopulation densityFile formatComputer animation
19:23
MassDifferent (Kate Ryan album)Point (geometry)MereologyPopulation densityMassSlide ruleState of matterComputer animation
20:33
MassArithmetic meanMereologyGravitationFigurate numberPopulation densityElectronic visual displayAreaType theoryRhombusSlide ruleCoordinate system
21:21
Local GroupArithmetic meanMeasurementGeometrySlide ruleMathematical analysisResultantStatement (computer science)Shared memoryNumeral (linguistics)NumberAverageProcess (computing)Semiconductor memoryCartesian coordinate systemGroup actionSequelComputer animation
22:21
Shift operatorNatural numberArithmetic meanRight angleComputer animation
22:50
Execution unitMaxima and minimaNormed vector spaceDuality (mathematics)MiniDiscForm (programming)Right angleSlide ruleGravitationDistanceSoftware developerWord
24:23
Shift operatorNatural numberState of matterStatisticsSource codeElectric currentNumberAverageBit rateCalculationComa BerenicesDifferent (Kate Ryan album)CalculationTwitterLine (geometry)Water vaporCurveMassComputer animationLecture/Conference
25:23
EmulationCalculationCartesian coordinate systemShift operatorMaizeDirection (geometry)Nichtlineares GleichungssystemSphereShift operatorGoodness of fitNatural numberShared memoryComputer animation
26:51
QuicksortWordSource codeStatisticsComputer animation
27:23
Source codeComa BerenicesCross-correlationMathematical analysisMathematicsDifferent (Kate Ryan album)Point (geometry)Term (mathematics)Multiplication signCASE <Informatik>Set (mathematics)Traffic reportingRight angleShift operatorNatural numberService (economics)Uniform boundedness principleData storage devicePresentation of a groupArmBit rateComputer animation
30:58
Presentation of a groupControl flowLecture/Conference
Transcript: English(auto-generated)
00:07
We've got three topics lined up. Mine is the first one, so I'll probably get right into it. And I'll be talking about the rise of China with regards to global economic activity.
00:22
And that's what I had a closer look at. And what I sort of tried to understand is how the trade flows changed in recent history. Recent history is basically the years 2000 to 2018. That's what I had data available for.
00:44
And this is all trade data based on the Harmonized Commodity Description and Coding System, the HS codes. That's an international standard which was developed by the World Customs Organization. And it's a six digit system. So for example I put down here 260, 600 stands for aluminium.
01:07
So that's how countries report. Some are more precise, some are not so precise about their reporting. So it really depends on the country as well, how precise they are. The data came in a tabular format, CSV. It had a few gigabytes so altogether they added up to around 28 million bilateral datasets.
01:35
So it's not only China, it's the whole global trade data that I had available.
01:42
And what I tried to do is, well of course first you put it in a database and then you try to make sense of it. So that's what I tried to do. I tried to move from exploratory to explanatory data analysis. And that's what I'm going to present here in the next few slides.
02:00
Because it's a software conference, right? So I'm going to talk a bit about the software environment that I used. And that's probably familiar to most because it's an open source conference so I completely relied on open source modules. So there's a Postgres database with the PostGIS extension.
02:21
And for the data analysis I used a Jupyter notebook. That's probably familiar to some of you in this room. So that's quite a neat tool to analyse data. And it broadly has the charting library and of course some other Python modules.
02:42
NumPy, SciPy to analyse the data. And for the geospatial visualisation I used QGIS and Leaflet. And I'm going to show you an animation with Leaflet in just a bit towards the end of the session. So this is just a brief look at the Jupyter notebook.
03:08
That's how I structured the data always. You have a few parameters, right? In the first line it's always a method to structure the data. You have the commodity as the first parameter.
03:23
Then you can choose between volume or value. It's always reported in terms of volume and value. And then you can have a look at the export or import. And you can have a closer look at the 10. Well, you can supply the number of leading countries.
03:42
So 10 stands for the 10 biggest countries you want to look at. And this is here, the broadly charting library. I used to use Matplotlib a lot. But I found it actually increasingly better to change. Because broadly, for example, it's much more interactive.
04:03
You have the legend entries here. Where's the pointer? Never mind. You have the legend entries. Oh, thank you very much.
04:22
Here, so this is all interactive. So if you want to have a closer look at a certain country, if you double click it, then the chart will change and only display the country that you've chosen. And you can turn on and off all of those countries individually. So that's something I really can highly recommend.
04:42
And that's why I actually use that as my standard library in these days. So what I do in Jupyter Notebook is always the same. I sort of leave the data in the database. There's no redundant data you have to create.
05:02
So it all remains within the database. But you retrieve it in the Pandas data frame. Which is here in this case. You retrieve a stored procedure from the database. And that's put into a Pandas data frame.
05:24
And in the second goal you just visualize it by supplying the data frame here in this function. And then while you have the title and the x-axis, and if you want to have it stacked or not, that's all the parameters you can code in here.
05:43
So that's quite a neat tool. Well, as I said, you have 28 million datasets. You don't have to process individually. You can just retrieve it one by one. Or you can filter it by the data that you want to look at. And that's what I'm going to concentrate on now.
06:03
Back to China. I have some macroeconomic data that I also visualized. So this data is from the World Bank. And in 1990 China started rank 11. So as the 11th biggest economy.
06:21
And they have improved significantly. So in absolute terms they are the second largest economy today on the planet. After the States. But this case here shows the GDP adjusted for purchasing power. And in this sense China is already the biggest economy on earth for the year 2017.
06:50
I've used 2018 as an example. So this is the increase just for China. So they've done a tremendous effort in the past 30 years.
07:01
They've taken 700 million people out of poverty since 1990. So they've really done a great job. If we look at the GDP growth. We can see that China is also. Well it has in the 90s it had a few years where two digit growth numbers.
07:25
So that's also a tremendous effort. More recently it sort of drops down a bit. But it's still among the fastest growing countries in the world. The other countries here.
07:42
India and Indonesia they are also growing quite rapidly. And that's something you will also see. When I show you the animation of the shift of trade powers. That actually the Asian countries are increasingly dominating the global market. With regards to natural resources.
08:03
Because that's what the trade data is all about. My institution because I'm from the Federal Institute for Geosciences and Natural Resources.
08:22
So we actually closely monitor mineral resources and metals. And I had a look at what China is the world leader at. And you can see this. This is quite a long list of natural resources. That China ranks number one.
08:42
So tungsten, mercury, rare earths, antimony, magnesite. The list goes on and on. Crefite, wollastonite, lead, manganese, barite. This is just the chart for China being number one in extraction.
09:01
So that goes on for number two and number three as well. So of course they also dominate the market in other natural resources. But with regards to number one positions. That's actually also quite a long list.
09:21
Same for refining. Refining is also why China is also the dominating leader worldwide. So rare earths is what they are known for.
09:45
There's the Bayan Obo mining city in China which has deposits. And they supply almost 50% of the world market with rare earths.
10:02
But also magnesium, indium, cobalt, aluminum, tin, zinc, lead, copper. So it's also quite a long list. They are also top leader in refining. In the year 2017. So with regards to natural resources. Definitely a very dominant player.
10:20
Or the most dominant player. And if we look at the aggregated data. You can see that China here has been quite an important player for most of the time. Only second to the United States in the year 2000.
10:42
But then taking over. With regards to exports. With regards to imports. The increase is much heavier. So that's an indication that the industrialization of China is moving rapidly.
11:02
So maybe number six here in 2000. And the average import share of all commodities that I looked at. They've increased their market share. And are now almost at 14% globally. So that's quite a huge increase over the past, well, almost 20 years.
11:32
And I've got two case studies looking at cobalt. What you see on the left is the whole import mix of China in 2017.
11:44
We've got a scatter plot here. That means those are pretty much the commodities China imports. And on the x-axis you see the country risk. So meaning where those imports come from.
12:03
If it's below zero or below zero. Below zero that means the mix of imports where the commodity comes from is from volatile countries. That's why it's also red here. That means where they have to import the natural resources from is from volatile countries.
12:24
As opposed for example to these ones here. They are from relatively stable countries. The y-axis is the Hirschman index. That's index of market concentration. So the higher the number is on that one, the more concentrated the market is internationally.
12:43
So if you end up here, that means it's quite a risky environment. As opposed to these ones. That means they are from stable countries and have a diversified market. If you have a highly concentrated market obviously you have increased risks of supply shortages or price risks.
13:09
Because as a monopolist you can always set a price. So that's why a high market concentration is also in many cases not really a good thing.
13:22
And cobalt ranks up as one of the riskiest when looking at imports. And that's what you can see here. China has been increasing its global share throughout the years to a very dominant player while the others have lost.
13:40
And this is because most of the global extraction of cobalt happens in the copper cobalt belt of Katanga which is in the Democratic Republic of Congo.
14:01
And they supply more than 60% of global cobalt. And also of course China's imports heavily rely on a country. So that's something that can negatively affect the market.
14:26
Same for tin. This is another example which has increasing risks. It's up here. I have a Hirschman index of 7,500.
14:40
Everything above 2,500 is considered to be risky. Or considered to be highly concentrated. And tin has almost 7,500. And also China really dominating the global import market. Because it's part of their industrialization process.
15:04
So they have a huge hunger for tin as well. And most of that tin comes from Myanmar. It's almost a monopoly. Almost 99% which is from Myanmar.
15:24
And that poses risks on the market for China. Generally if you look at the worldwide development, this is the annual growth rate that I calculated for those data sets that I had available.
15:45
And China is not the fastest growing. I found Philippines was growing by more than 14% every year since the year 2000. But also China, India, definitely the fastest growing economies are in China.
16:06
Whereas the traditional hubs of North America and Europe. Well they actually had a loss in natural resources. Which obviously doesn't mean that they have a loss in GDP.
16:21
Because I'm only looking at natural resources. But with regards to industrialization some of these countries went down. They probably grew in other areas like services. But my idea, and that's why I'm coming back to the methodology, was to calculate a spatial mean.
16:48
So I wanted to quantify how the power shift happens globally. And that's why with these 28 million data sets I calculated a spatial mean.
17:02
So I have one point each year which is the center of gravity which is calculated from each country's contribution to the trade data. So let's say it's around 1200 traded commodities and each country takes part within a few commodities.
17:24
And that proportion I calculated to see where that center of gravity falls each year and to see where it moves to. And what I also did is center of gravity, well you have different center of gravities right?
17:44
I mean it can be centered with regards to land mass. That can be calculated for a country. What I used as the center of gravity is actually for each country where most of the people live. So what I did is I downloaded the population density raster from the socioeconomic data and application center CDAC.
18:12
And I clipped each country as a raster and calculated with r the center of gravity with regards to population density and not land mass.
18:26
And that's a huge difference. We're going to see that in the next slides. Did I? Excellent. Sorry about that.
18:49
That's what we're going to see here. Well this is just the code I put down just to prove that I didn't get any apprentices to do slave work manually. So that's the scripts that I used. It's basically a for loop and they take the polygon and cut the raster.
19:08
And put it in a post just raster table format. And with r I go through another for loop and calculate the center of gravity.
19:20
So that gives us an adjusted point feature data set which is adjusted for population density and not mass. And I think you can see that in so many countries that there's a huge difference.
19:41
If we look at Canada for example it falls within the United States. Actually it crosses the border which does make sense because most of the people in Canada live on the southern side. And also for the states it moves more than, no it moves almost 2000 kilometers because you have the land mass of Alaska up here.
20:05
So that's where the centroid for the land mass falls. And with population adjusted it falls 2000 southeast of the other centroid. And some of the biggest countries they all have completely different population centers than land mass centers.
20:30
So next slide is the same for the Asian countries. You can see that also China, I can't see that figure but it's also
20:42
more than 1000 kilometers away because you have the Himalayas here and the Gobi Desert. And the white parts here is the highly densely populated areas. So that means of course that center of gravity is now here which makes sense
21:01
because most of the population here, the white parts, they live more towards the coast. And I thought that's a more accurate way to show the spatial mean. So that was my methodology.
21:23
This is another slide about what the spatial mean actually is. So it is actually, you're looking at two dimensional XY, longitude and latitude. And you weigh that coordinate by the share of the commodity they traded in.
21:45
And that gives the final result in SQL, that statement here after some processing. It's actually an aspatial analysis so far because X is a numeric value, Y is a numeric value.
22:02
So all it does, it builds the average. It's an aspatial group by statement. And let's have a look at what the result is.
22:53
That's what I found. That's how the spatial mean moves in 18 years.
23:00
So you can see that the pull towards China is tremendous. In the year 2000, the center of gravity was here just off Sicily. And then it moves in the Mediterranean right into Syria in 2018.
23:25
And overall it moves 2006 kilometers. And you can see that the distances from the developed world in North America or Germany,
23:41
that increases from year to year. And on the contrary, the Asian countries, they actually decrease the distance. So that means that global trade increasingly happens in Asian markets.
24:18
The slides.
24:39
And as one of the last slides, I'm almost done.
24:42
I've compared some of the differences in calculation just to see whether that trend comes really from China or from other countries. So this one is the trend without China. Way more westward and not as pronounced as the other one.
25:06
And these differences, these two lines is world population adjusted or land mass adjusted. So what I figured is that actually the population adjustment pretty much both curves moves from north to south.
25:25
Which probably does make sense because less people live in the northern hemisphere and more towards the equator. So most of the shift actually happens in a southward direction if you adjusted for population.
25:42
And this is the imports, this is exports by the way. So what I can definitely say is that the pull of China really changes the whole picture. So if you take out China out of the equation, other Asian countries also pull but China definitely pulls most.
26:02
Yeah, that's the last slide. So it's the, in the past 20 years China has been tremendously growing.
26:22
So they are a dominating player in extraction, refining and trade of natural resources and also in intermediate goods. And traditional hubs like North America and Europe increasingly do share with regards to natural resources.
26:44
And yeah, that's pretty much it. Thanks for your attention and I'm up for some questions if you have any. Thank you. Any questions?
27:03
Thanks. Did you also compare the statistics with other data sources like the National Statistics Bureau of China? What sort of data, what data exactly you mean? The trade data that you were using.
27:20
No, I did not know. Which data are you referring to, the census data of China itself? I mean the trade data with other countries. What do you mean with other countries? Well the trade data with natural resources, with all the other countries, so most of the data that you were investigating here.
27:49
Yeah, well it's a global data set, right? I mean there's not only data for China, so there's data for all countries in the world.
28:02
So it's bilateral trade data. And there's a huge difference how they report. So for example, that cobalt example that I showed you where China imports most of it from Congo. If you look at Congo itself and what they report to the customs, it's zero.
28:25
So they don't report anything. So that's a big change. But the data set itself is from one source, it's a proprietary source. And that's the only one I had available. So there's other data sources around like the UN trade com, or com trade it is.
28:47
But I use this one for the analysis because I think it's a bit better.
29:02
Thank you for your presentation. Some people say that China's success in the economy relies on their working age population. And in 10 years time or 20 years time, they're facing an aging society in a very fast pace.
29:25
Because even though they lifted the one child policy, the population is not so much growing. And in 30 or 40 years time, some people say that the economic center of the world will shift to Africa. Because there is a big, large population growth. And I would like to see how the working age population or population bonus has been affected by that special shift in the economic center.
29:50
How it will in the future? In the future? Or in the past? You are thinking about the correlation between...
30:04
Right, right, right. Well, that is definitely an interesting point. And it's probably a very valid point. Because in many cases like Europe and North America it happens already, right?
30:20
And that might also partly explain why those countries go down with regards to industrialization. But it also means, as I said, this is natural resources so I didn't have any data available for services. So in terms of overall GDP, it probably looks a bit different too.
30:44
Even though I looked at it and there's a pull towards China too. But that's something you probably could do in a statistical analysis. And see the correlation between population growth and GDP growth.
31:03
You are welcome. Any more questions? Then we can close the first session of myself and have a few minutes for break. And moving rooms.
31:21
And the next presentation is at 9.30 in this room. Thank you very much.