Retired computer programmer 網誌


John Tsang, what you are hiding?

John Tsang, what you are hiding?


Summary: While we fiercely debate the 0.75 billion of extra CSSA payout per the 7-year residency retraction for one-way immigrants, we let go a 2 trillion government pensions exposure issue. Which one should we worry now?

[Post update (Jan 10, 2014): After my original posting on Jan 6, 2014, at a re-read, I've noted some errors in my calculation of the 4T exposure. It's now revised to 2T. Also, my assumption of a dwindling government revenue in a stagflation scenario appears to be flawed, as while the real GDP of the economy might contract, the nominal GDP doesn't. A better comparison would be to project the share of pensions expenditure in total spending rather than a comparison of the present value spending against a future inflated pensions sum. Since this article has been published, I leave it here for the record. I will post a second analysis when done. At the mean time, corrections below are annotated accordingly.]

In discussions about social welfare in HK, you usually see this chart, hailing the effectiveness of Donald Tsang's 7-year residency rider for one-way immigrants introduced in 2004, on curbing the expansion of CSSA spending:

(source: Statistics on Comprehensive Social Security Assistance Scheme 2001 to 2011)

However, you seldom see this chart (in the same source, just 2 pages apart):

You also seldom see this chart, which shows the HK GDP annual growth in the same period:

So, is it the intensified SFS measures introduced in Jun 2003 and the economic recovery in the same period that curb the CSSA spending, or is it solely yielded by the 7-year rider?

Also, you may see statistics in article like this, which has a part that highlights one-way immigrants attaining tertiary education. It rose from 5% in 1998 to 10% in 2003, then to 14% in 2012. So, it's kind of hinting that the 7-year rider has a positive filtering effect by education level on one-way immigrants.

However, the author forgot to mention the UNESCO statistics, which shows that tertiary enrolment in China rose from 3% in 1991, to 24% in 2011.

I do not intend to debate whether the 0.75B (billion) was under-estimated or over-estimated by the judges. What I want to say is that the 7-year rider is at best a lukewarm “success” on curbing the CSSA spending, and the retraction of it, will at most exhibit a muted impact upcoming.

Let's take a look of the overall distribution of government spending:

(source: stacked chart compiled from section 9.5, Hong Kong Annual Digest of Statistics 2013 Edition, p. 247)

So, at least, your government money is not all spent on welfare. It's just sitting humbly at stack 2 from the top on the chart. There are other spending which exists mostly as implicit subsidies to everyone.

Let's take a step further:

(source: stacked chart compiled from Table 194: Government Spending – General Revenue Account and Funds)

The chart is based on another table on government spending which has the personnel expenses stripped out to their own categories. It shows the subventions actually paid out. Now, the welfare stack seems even more irrelevant. [post note: category Social Welfare on Table 194 appears to have NOT included CSSA and understated the total welfare subventions. For 2012-13 as an example, if CSSA was included, the Social Welfare stack shown should be about 3 times as thick on the chart. I am not sure where CSSA was put in Table 194, perhaps in Other Charges, which is unusually large.]

All of the above are not the main focus of this writing. So far, I just try to give you a perspective on how “heavy” welfare is placed on our government spending.

Now, here comes the real meat. From the same Table 194, we narrow down to 3 recurrent expenses on personnel and show it in a percent stacked chart-

Do you see how pensions is eating into the overall personnel spending? As of 2012-2013, pensions spending is 21.8B. As of this point, not a staggering concern compares to the total operating expenditure of 303B. However, read on.

If you take a look of the same data in a regular column chart:

You will find that the some 20B of pensions we paid out last year is about the same as the 20B of Personnel Emoluments we paid 20 years ago. This seems logical because assuming an average working age of 40 for the civil servants, those who worked in 1990 is about to retire in 2010, and the government will pay them continuously until the retired die. While the exact pension terms differ (they are not paid the full monthly salary as they would have in service, and there is also a part in lump sum payment), but at a high level in terms of total pension payout, it appears to be linked in this way

For those who are not familiar with pension schemes, here is a brief explanation. In the old days, we have defined benefits retirement plan in general called pension scheme. Disregarding how much you and your employer contribute to the fund, you are always guaranteed for a fixed income after retirement discounted based on the last salary in service. This retirement payout will continue and inflation adjusted until the day you decease. For our government's pension scheme, here is the link.

Defined benefits scheme was popular because from mid 80's to year 2000, we were having one of the best times in global economy that was later coined as the Great Moderation. GDP grew, inflation benign, and global stock markets were on a moonshot (S&P 500 rose from 200 to 1500 points, a more than 7-fold gain!). Pension funds that had a portion invested in stock market were always comfortable to make payouts. Further, the demographic window was wide open with the expansion of working population more than enough to cover the retired.

However, things have changed since last decade. The demographic window is closing and stock markets stagnant. No more defined benefits scheme but more on defined contribution scheme, which is what you see in our MPF. You and your employer contribute but you can get only what you've paid for - there is no guaranteed income after retirement.

To be fair, our government also switched to MPF for the newly hired since 2000 and thus at least, the volume of current and to-be pensioners is no longer growing. However, the risk continues as you won't see a stabilized pensions expenditure until around 20 years after 2000. After 2020, the burden will carry on for probably another 25 years (based on average life expectancy of age 85).

This is a risk because we see 2010 pensions expenditure matching 1990 emoluments likely due to a deflation cycle in the period, and thus curbed the salary increment:

So, if next 20 years (from 2000) period is the upcoming decades present a nightmarish stagflation scenario, we might see the pensions growing from the projected 50B/yr in 2020, to 80B/yr in 2045 expenditure doubling to 100B/year from the about 50B/year in 2000 emoluments (assuming an annual master pay scale increment of 4%, 1.04^20=2.191) while government revenue dwindles.

If for the next 25 years after 2020, stagflation continues, then the 100B/year might further balloon to 160B/year (assuming an annual inflation rate of 2%, 1.02^25=1.641 from 2020 to 2045).

So, this is what John Tsang is worried about. It's not your welfare. It's also not your medical and education (if the government dares to downsize schools in view of a foreseeable influx of toddlers from mainland, what makes you think they wouldn't do the same for medical and education expenditure whenever revenue runs dry?).

It's about the more than 4T close to 2T (trillion) of total pensions exposure from now to 2045!
(2013-2020, from 60B/yr to 100B/yr20B/yr to 50B/yr, straight line interpolation average 8035B/yr, 17 years total 1.36T245B
2020-2045, from 1050B/yr to 1680B/yr, average 13065B/yr, 25 years total 3.251.625T)

So, finally, we have our own 4 2T “stimulus” matching the mainland back in 2009.

And you tell me to worry about the 0.75B/year or perhaps, make it to 10B/year, extra welfare due to the 7-year residency retraction, which requires about 42 centuries to match up to 42T? You must be kidding!

Am I blind guessing John? Apparently not. It was Regina Ip who raised the eyebrow first. She is rightfully concerned because well, she is on pension too. (Regina's 2012 query to the government on pensions)

Separately, Regina also publicly mentioned this issue in 2013.

Am I exaggerating? Perhaps, but as you can see from the Regina's query link above, the government didn't provide too much details on the subject except that as usual, they spent a lot of ink on why they can't give further details. So, what can I do except a ballpark estimates? I welcome anyone who has more expertise on the subject of pension to correct me.

From a global perspective, for those who think that it's welfare that brings the US fiscal budget to its knees, read this.

If John is really so worried about our fiscal budget, he should, like Ford and Detroit, immediately start the discussion with civil servants and retirees on pension cuts/buyout. I am not talking about a total cut of 42T because the civil servants deserve their share of retirement benefits (although currently, MPF allows bankrupt companies to cut fully their part of balance), but an adjusted scheme which perhaps linked not only to last salary and inflation, but also to the growth rate of revenue of the government. If government's income is good, we grant the terms. If income is bad, civil servants have to sacrifice a bit. Is that fair?

Of course, John won't listen. You can imagine the heated debate and struggle on working out an adjustment for pensions with the union. Besides, why would he work so hard to, in the end, cut his own benefits? So, he found an easy exit – just hint that it's the one-way immigrants who ate your lunch, and everyone starts crazily chasing after them for revenge, even though it's the government who created the resources conflict from the very beginning by downsizing schools, turning a blind eye to crowded hospital since 2000, and contracting the ratio of public housing expenditure for the last decade (perhaps to make space for pensions?). Forget about the 42T issue - 0.75B is more important. I guess CFOs around the world must now be so envy for the job of Financial Secretary in HK, since they first time in history, can tell shareholders that 0.75B reduction in company expenses is much more important than containing a 42T risk exposure, and shareholders buy it. What a dream.

(source: Housing in Figures 2013, Hong Kong Housing Authority)

For John, less time for work, more time for red wine, what's not to like? Why the fuss to talk about pensions?