<收入及工時按年統計調查報告 : Some comments – Part 4>

Part 1

Part 2

Part 3

The last issue regarding the study : the data collection period is April to June 2009 (i.e. 2nd quarter of 2009). Why is it important?

Given this is an essential input for SMW, and SMW is so important for the economy, we must ensure the data are relevant, representative and free from bias.

So what happened in Q2 of 2009? Every part of the world, including Hong Kong, was suffering from the financial tsunami. Economic outlook was dim at that time, with cool off in wide range of economic activities and rising unemployment. 

According to the statistics from Census and Statistics Department, wages declined for almost all sectors (ranging from -0.1% to -3.5%; see here). The unemployment rate was also increasing – Indeed, the 5.4% unemployment rate in 2009Q2 was the highest during the crisis.

There is little more than dangerous than basing your policy and advice on just one snapshot of the economy, like this report (covering only 1 quarter). The information so collected is not reliable, nor relevant. Suppose that with the gradual pick up in economy since the second half of 2009, wages across all sector, gender and education attainment rose. If you set your minimum wage basing on the previous set of wage data, the minimum wage will only cover less people than expected.

Of course, the dis-employment impact will be less. It is one thing that there is less dis-employment, but quite another that you set it incorrectly.

Do they make it up deliberately? I do not think so, because the lead time for survey and data collection is very long. You probably take half a year to plan to have such a study, and another to design the relevant survey etc.

Put it in another way : If the government is so good at fixing the survey period at the time when the financial tsunami was attacking the economy at the hardest, our government will not be in such degeneration mode.

<收入及工時按年統計調查報告 : Some comments – Part 3>

Part 1

Part 2

The implications of higher actual hours of work

I have assumed in the previous part that across all industries, occupations and gender (and other aspects) the actual working hours are higher than the one collected by the Census and Statistics Department under that definition.

If this is the case, the actual hourly wage must necessarily be lower than the report one (Median : $58.5 per hour).

But my rough idea is that, unpaid over-time is a very prominent and widespread phenomenon in the following industries – only mentioning those covered by the report (figures in the brackets are their median hourly wage) :

Incidentally, the hourly wages in 3 of these 4 industries are the highest in all industries surveyed. That means, in these industries, the hourly wage basing on actual working hours are lower than the report one. In turn, the overall median hourly wage, should also be lower, even assuming the actual working hours for other industries are the same as reported under the definition of “contractual hour”.

More importantly, since we are adopting a “median” hour of work, no matter how the numbers below the median get lower, they have no impact on the median.

Let’s consider this example : 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11

Median : 6

Though not perfectly correct, the above example means, in lowering the median, it is far more effective to work on the numbers above the median (and reversely, if you want to boost your median, all you need to do is to work on the numbers below the median).

In other words, if we were to use actual working hours only for those highly paid industries, we should possibly see a substantially lower hourly wage for all industries as a whole.

—-

But is it fair to say those highly paid industries have lower hourly wage, basing on actual hour of works? It depends.

The hourly wage concept for such industries are, unfortunately, useless. Why? Because employee i in these industries work throughout the year, in the expectation that they could receive discretionary bonus/ performance-based bonus at the end of the year (which are both excluded in the study).

It is very debatable whether discretionary bonus should be counted. But such exclusion should be borne in mind whether interpreting the hourly wage for these high-paid industries.

<收入及工時按年統計調查報告 : Some comments>

Part 1

Note : I am NOT against this report. Amongst all government departments, Census and Statistics Department is one of the most reliable. Not because it is objective per se, but mainly because it has to follow many international standards for compiling the statistics, so that there is little room for them to “mauipulate” things around.

But I AM against Minimum Wage.

The point of this whole entry is to highlight some issues that are too easy to be missed. I will also cover some of their implications on the findings.

1.  Hours of work

— I read the report for a number of times more. Still cannot find any table showing the number of working hours.

— But I manage to find the definition on the hours of work :

So, you know what I am going to say, right? Anyone working in Hong Kong, except civil servants (who are excluded from the report’s coverage) has to work overtime, unpaid. The issue is not “yes or no”, but “more or less”.

Contractual hour of employment, therefore, must be lower the hour directed by the employer. Furthermore, effective or actual working hours are even larger : You have to meet this deadline, so, you don’t really need employers’ direction to work over time. There is no agreement, nor being written in the contract of employment. Working over-time, on an unpaid basis, is just automatic.

One prominent example is the junior worker in the so-called Big-4 account firms. Look at some discussion in 高登 here.

Since the hourly wage is simply wage / hours of work, using the contractual hours of employment or the agreed hours or the directed hours, which are all lower than effective working hours, will give you a higher “nominal” hourly wage.

In other words, were the higher number of actual hours of work be used, the hourly wage so computed will be lower.

— So the next question is : Should the actual hours be used? Let me ask you one very simple question :

I can only give you a rough idea. The answer to this simple question is not so simple, and inevitably involves guess, range-estimates, and probably other methods that are less reliable than, at least, the contractual hours of work. So I am of the view that the way of defining hours of work (through contractual working or agreed hours, which are more reliable and can sometimes be proved by evidence) is acceptable.

But, the difference against actual wage should be borne in mind (which is not in the report), and I believe for comparison purpose, a set of hourly wage data (to the simplest level) basing on actual working hour should be compiled (which is also absent in the report).

—–

There are several more parts to go. Given that this site is a Twitter-like xanga, I try to keep post as short as possible.

<收入及工時按年統計調查報告>

You can download the report here. 

Media already has some headline coverage (e.g. MingPao) :

I do not have the time to find out interesting observations, yet.

But there is already one funny thing about this report :

Stay attuned to more updates ahead.

<Applied science>

This is from MingPao :

What should the person(s) mistakenly delete the message do?

Of course, he/she should hire the team from EMSD. They should at the end come up something like this :

Wao, this is very amazing, and more importantly, useful.

<Crap>

EMSD finally released its report on the LPG incidence (final report here, appendix here)

This is the major conclusion (p.53 of the report) :

7.5 專案小組沒有證據確定中石化該8 次懷疑未能按照氣庫工作程序進行的「扽缸」運作，直接引起今次死火事件，但專案小組認為懷疑未按程序進行的「扽缸」，有機會影響石油氣的品質。

Of course they cannot find evidence. But the relationship can be less than sufficient “扽缸”, then LPG quality being affected, and causing 死火.

The whole report is pseudo-scientific, without many piece of seemingly useful information. 

This is the only possible piece of evidence that is likely to cause the incidence. So what would you expect?

1. Relevant records of 扽缸 by other oil companies. This is simple.

If we discover that other oil companies have sufficient time for 扽缸, then this gives more evidence and confidence that insufficient time for 扽缸 may cause the incidence. If NOT, this can prove 中石化 more 清白, as the insufficient 扽缸 and 死火 are just correlation (A happens, then B happens; but A does not cause B), instead of causality (A happens, then B happens; A does cause B).

2. Some hypothetical 扽缸 with less than sufficient hours. Then test the quality of LPG, use it to test with it will lead to 死火. This is not so simple.

This can even provide strong evidence to prove or disprove 中石化.

Yet, basically, so conclusion is worse than no conclusion. This is a shame, because such committee has put no real effort in finding the true cause, and more importantly, getting the appropriate way to avoid similar problems in the future.

One good thing : This report amazes supporters of the conspiracy theory that our Government dares not to disclose the failure / fault of any Mainland enterprise.

(earlier entry on 死火)

Suddenly, for no reason, I recall the story of 陳浩南 and 山雞 in the comic 古惑仔. Not the movies, in which both eventually turned themselves back to brothers.

After all the tears and blood throughout the years, 陳浩南 and 山雞 eventually become 宿敵 for the rest of their lives.

Love, 手足 turned into hate and 宿敵.

Everything will not end until the death for both of them.

<Deloitte’s Football Money League 2010>

Soccer fans are strongly encourage to take a look on this interesting report. Click here for the homepage. You can also download the full reports (this year and in previous years) for free by simple registration.

Regarding the impact of financial tsunami : UK, Spain and Italy are hardly hit by the crisis (e.g. Spain is the S in the PIGS – Portugal, Italy, Greece and Spain).  Interestingly, these are the countries where the traditional top soccer teams locate (Manchester United, Chelsea in UK; Read Madrid, Barce in Spain; AC or Inter-Milan in Italy).

Surprisingly, there seems to be no apparent impact on revenue on the soccer teams. If you look at the report, non UK teams will report higher revenue. Meanwhile, revenue in euro terms for UK teams was either stagnant (e.g. Manchester United) or declining (e.g. Liverpool). But this is purely due to foreign exchange conversion – Pound dropped against Euro during the crisis, so when you translate UK team’s revenue (in Pound) into Euro, you get a lower number. 

In other words, if you look at UK team’s revenue, Liverpool and Manchester’ revenue is indeed rising. But Chelsea’s still declined.

In terms of capacity utilization (how much your stadium is filled), at the end of the report, there is an average 2% point drop for these 20 top soccer teams. Again, there seems to be no apparent negative impact.

The report does not say whether these teams are making profit. The cost for running a team is very huge, particularly on the players’ payroll. This is why 12 NBA teams (out of a total of 30) were losing money last year. Interestingly, in NBA, teams taking huge revenue or making money are not necessary the better teams (except the Lakers). In the top 5 earning teams, Chicago Bulls, NY Knicks, Houston and Pistons are at best mediocre.

Also read this excellent piece on NBA’s revenue shortfall from Bill Simmons.

<Interesting>

Despite the improving economic condition, ruling parties and their leaders in various parts of the world are losing popularity to the opposing parties. This seems to be an all-of-a-sudden disease.

Taiwan : 民調滿意度僅 29%　馬英九民望大插水 (Mar-1)

Japan : 親華鳩山招攬挽聲望 (Feb-24)

UK : 內閣辦公室 7%人受魚肉, 唐寧街高級職員欺凌同事 (Feb-24)

In these economies, you prefer to be the opposing, rather than the ruling, parties. Ridiculously, whoever comes to power in Japan is an end to the power as well. Who still remember 鳩山’s government was quite popular six months ago?

The same goes for Taiwan. KMD just lost some important by elections. Who still remembers 小馬哥?

Where are the oasis? Of course, our Motherland.