Archive for the ‘budget’ Category

Miaoli County government runs out of money

July 10, 2015

The Miaoli County government is completely broke. Not broke as in, it can’t afford a fancy National Day celebration and it’ll have to make do with a modest one. No, broke as in, it can’t pay basic expenses on time. Civil servants didn’t get their paychecks as scheduled this week. The county government is currently NT800m in the red, and that’s after some creative accounting is already figured in. One prominent media figure is comparing the situation to the current Greek crisis. It’s not an entirely crazy analogy.

How did this happen? As soon as the new county magistrate took office and got a peek at the county finances early this year, he immediately started screaming that the previous administration, led by Liu Cheng-hung 劉政鴻, had spent the county into financial disaster. This is not a partisan ploy. Both the prior and the present magistrates are KMT members, though they are from different local factions. The KMT owns 100% of this calamity. Their guy borrowed way too much, spent the money irresponsibly, and the central government never stepped in to slow him down. They didn’t even complain.

It should have been obvious to the technocrats that something was going wrong. I’m a novice when it comes to finances, and even I can see a problem in the numbers.

(Note: The numbers are hard to get straight. The newspaper report says Miaoli has a total debt of NT64.8b. I can’t find that number in official documents. The numbers I have found are NT48.7b in May 2015 or NT49.1b in 2013. I’m going to be using the document with the 2013 number for the rest of this post, so keep in mind that the actual debt might be 30% higher. Even if I am using the wrong year or making some other minor mistake, the broad story is unmistakable. Miaoli is not like everywhere else.)

Miaoli County debt went from NT16.2b in 2005, the year before Liu took office, to NT49.1b in 2013. That seems like a big increase. However, we need some context. Regulations governing how much debt local governments were allowed to incur were revised, so everyone’s debt went up. Further, we shouldn’t compare Maioli with Taipei. Direct municipalities run under different rules and have very different revenue streams. In this post, I will only look at the eleven county governments in Taiwan. That is, I’m not looking at Taichung or Tainan Counties (which were upgraded to direct municipalities), Keelung or Hsinchu Cities (which, as cities, don’t have the same sorts of revenue or expenditure profiles as a rural county like Miaoli), Kinmen (which is mostly funded by the alcohol factory) or Matsu (which gets nearly all of its budget directly from central government subsidies). The eleven counties in this table (especially the first eight) face roughly comparable fiscal challenges.

2005 debt 2013 debt % increase
(billion NT) (billion NT)
Yilan 20.32 25.28 24.4%
Hsinchu 22.66 27.76 22.5%
Miaoli 16.23 49.05 202.3%
Changhua 14.35 33.05 130.3%
Nantou 12.58 17.90 42.3%
Yunlin 23.58 31.99 35.7%
Chiayi 19.57 28.69 46.6%
Pingtung 17.92 28.02 56.4%
Taitung 4.52 9.08 100.9%
Hualien 8.76 12.41 41.6%
Penghu 1.12 2.18 93.7%
6 KMT counties 71.46 139.01 94.5%
3 DPP counties 61.06 88.70 45.3%

Compared to the other ten counties, Miaoli stands out. Miaoli’s debt increased by over 200%, which is another way of saying it more than tripled. The next biggest increase is from Changhua, which increased by a mere 130%. Everyone’s debt went up, but Miaoli’s debt exploded.

Since I know everyone wants to make a party comparison, I’ve summed the totals for the six KMT governed counties (Hsinchu, Miaoli, Changhua, Nantou, Taitung, and Penghu) and the three DPP governed counties (Yunlin, Chiayi, and Pingtung). On the whole, the increase in the KMT group is roughly double the increase in the DPP group. Does this mean the DPP is the party of fiscal responsibility? Hold your horses there, Sonny. There’s more to this story.

Maybe Miaoli could handle the increased debt load. If Vanuatu and the USA both borrow a billion Euros, it will be a big burden for Vanuatu while the USA will barely notice it. We really need to know something about how much debt Miaoli is capable of carrying.

2005 revenues 2013 revenues % increase
(billion NT) (billion NT)
Yilan 17.20 18.62 8.2%
Hsinchu 19.60 24.62 25.6%
Miaoli 18.01 26.33 46.2%
Changhua 28.04 35.49 26.6%
Nantou 17.05 20.30 19.1%
Yunlin 19.01 27.79 46.2%
Chiayi 14.56 20.90 43.5%
Pingtung 24.76 30.58 23.5%
Taitung 9.21 12.63 37.1%
Hualien 13.38 17.34 29.6%
Penghu 6.81 8.05 18.2%
6 KMT counties 98.71 127.42 29.1%
3 DPP counties 58.32 79.26 35.9%

Miaoli has roughly the same population as Hsinchu, Nantou, and Chiayi, so you would expect those four to have similar revenues. In 2005, the first three were roughly similar, with Chiayi trailing behind. By 2013, Miaoli was outspending Nantou and Chiayi by quite a margin, and it was even ahead of fast-growing Hsinchu. Miaoli’s revenue stream increased by 46.2%, tied with Yunlin for the highest growth of any county. (Data on revenues and expenditures can be downloaded here.)

2005 expenditures 2013 expenditures % increase
(billion NT) (billion NT)
Yilan 18.83 18.62 -1.2%
Hsinchu 21.39 25.02 17.0%
Miaoli 18.14 26.33 45.1%
Changhua 32.40 39.84 23.0%
Nantou 19.50 20.30 4.1%
Yunlin 22.36 27.79 24.3%
Chiayi 18.88 21.40 13.4%
Pingtung 26.74 30.58 14.4%
Taitung 10.78 14.40 33.5%
Hualien 15.38 17.34 12.8%
Penghu 7.69 9.43 22.6%
6 KMT counties 109.92 135.33 23.1%
3 DPP counties 67.98 79.76 17.3%

However, Miaoli also led the way in spending growth. A lot more money came in, but all that money went right back out. Miaoli’s expenditures increased by 45.2%, just about the same percentage as its revenue growth. By contrast, Yunlin’s expenditures only increased by 24.3%. This brings us to the final table, the one that really matters.

2005 debt as a % of revenues 2013 debt as a % of revenues % increase
Yilan 118.1% 135.8% 15.0%
Hsinchu 115.7% 112.7% -2.5%
Miaoli 90.1% 186.3% 106.7%
Changhua 51.2% 93.1% 81.9%
Nantou 73.8% 88.2% 19.5%
Yunlin 124.0% 115.1% -7.2%
Chiayi 134.4% 137.3% 2.2%
Pingtung 72.4% 91.6% 26.6%
Taitung 49.1% 71.9% 46.5%
Hualien 65.5% 71.5% 9.3%
Penghu 16.5% 27.0% 63.9%
6 KMT counties 72.4% 109.1% 50.7%
3 DPP counties 104.7% 111.9% 6.9%

In 2005, Miaoli was firmly in the middle of the pack. In 2013, it was far, far more indebted than any other county. I don’t know what level of debt is sustainable, but judging by Miaoli’s inability to pay its bills, I’m guessing the magic number is somewhere below 186% of revenues. The current magistrate’s complaint, that Liu Cheng-hung’s administration spent the county treasury into a crisis, appears to be entirely reasonable.

What about the comparison between the KMT and DPP? It is tempting to see that the three DPP governed counties have only increased their debt load by an average of 6.9% and conclude that the DPP is much more responsible. However, I think that is far too simplistic. Sometimes borrowing is responsible. If the money is invested wisely, increased debt can set the stage for long-term prosperity. (Miaoli is said to have blown its money on extravagances such as elaborate fireworks shows and invitations to international celebrities such as Sarah Brightman. That probably wasn’t wise.) Moreover, if we are simply to look at the fiscal situations, we must consider responsibility for the starting points. In 2005, Yilan and Chiayi were two of the most indebted counties. The DPP had governed Yilan for 24 years, so it was completely responsible for the 2005 debt. While it had only governed Chiayi for four years, the DPP county magistrate had increased the debt by 28% in the previous three years. Similarly, we probably shouldn’t give Yunlin and Chiayi too much credit for keeping their debt growth low from 2005 to 2013 since the 2005 debt levels were already so high. I don’t think we should draw any broad conclusion from a simple table like this about the performance of the two parties.

However, given the current state of Miaoli’s finances, I’m pretty confident in concluding that the previous administration borrowed and spent irresponsibly. In every one of these tables, Miaoli is the extreme case. Why didn’t the Finance Ministry step in to investigate what the hell was going on? They should have been able to see the broad trends developing, and they should have had enough contextual information to know that Miaoli couldn’t sustain that debt. And where is the Control Yuan? Sorry, I got carried away. Everyone knows that the Control Yuan is only used to harass the other party, not to investigate actual government incompetence or malfeasance. Any real oversight will have to come from the voters.

Local debt

April 6, 2010

Here’s another reason it is good to be a direct municipality rather than an ordinary city or county.  Direct municipalities can take on more debt.  If I am reading the Public Debt Law 公共債務法 correctly, the current law says that Taipei City can take on up to 3.6% of the national GDP, Kaohsiung City can take on 1.8%, all the cities and counties put together can take on 2.0%, and all the townships put together can take on 0.6%.  Each county has a maximum of 45% of its annual income, and each township has a maximum of 25% of its annual income.  The Finance Ministry has a bill in the legislature that would change this so that direct municipalities and counties/cities could take on 250% and 70% of their annual revenues, respectively.

One might argue the merits of allowing local governments to pile up debt, but it seems to be irresistible to the actual politicians in charge.  I don’t have figures on how much debt each local government has amassed, but I’ve never read a story that didn’t say that the local government was either already maxed out or preparing to float more debt up to the legal limit.  I think it’s a fair bet that the incoming Xinbei City mayor will find it necessary to embark on a huge spending program.

KMT scores an own goal

April 6, 2010

I recently wrote about the allocation of centrally-funded tax revenues to local governments 中央統籌分配稅款.  This topic has resurfaced in a minor news story over the last week.

While interpolating Finance Minister Li Shude 李述德, KMT legislator Cai Zhengyuan 蔡正元 lashed out at DPP local governments for opposing ECFA.  He averred that local governments that opposed ECFA should not be allowed to get any of the centrally-funded tax revenues.  Lu Shude replied that Cai had read his mind (roughly translated).  The DPP immediately seized on this exchange, charging the KMT with political blackmail and no understanding of democratic politics or the relationship between local and central governments.  KMT secretary-general Jin Pucong 金浦聰 tried to damp down the controversy by suggesting that Cai did not speak for the party and perhaps should be more careful about what he said in public.  Cai responded indignantly that Jin didn’t know what he was talking about.

This was an enormous gaffe for the KMT.  Threatening to take away funds from local governments is a pretty good way to ensure that voters in those counties never vote for the KMT again.  It also violates principals of free democratic debate.  At the very least, one would hope that the KMT believed that ECFA will provide sufficient benefits that they could sell it to voters rather than simply threatening voters that they must accept it “or else.”  More generally, debate on public policy is supposed to be acceptable in democracies.  Various sides are supposed to be able to vigorously present their arguments.  It would be different if a local government were to actively violate some law passed by the central government, but, as far as I know, that has not happened.  ECFA has not even been signed yet, and local governments don’t have much power over international trade regimes.  At this point, the various local DPP governments are simply yelling that ECFA is a terrible idea.  In a democratic polity, that is not a cause for fiscal retribution.

From the KMT’s point of view, the worst thing about this flap may be that Cai and Li caused a stir for no useful reason.  The allocation of tax revenues is codified in law.  It is not a discretionary pool of income that the KMT can dole out according to its political needs.  In other words, there was never a chance that the KMT could actually carry through with this empty threat; the only effect is the backlash.  Shouldn’t the Finance Minister have known this?  Also, shouldn’t Cai have known that he was making a huge political error?  Li is a technocrat, so I’ll forgive him for not realizing what a political blunder he was committing by going along with Cai.  Cai Zhengyuan is a fairly savvy politician who is supposed to be good at this sort of thing.  He used to be the KMT party spokesman and the spokesman for the Ma presidential campaign, for crying out loud.

slicing the cake

March 27, 2010

Three days ago, in response to an interpolation question in the legislature, the Ministry of Finance announced estimates for next year’s allocation of centrally-funded tax revenues 中央統籌分配稅款 to local governments.  This is a huge pool of money; it accounts for somewhere between 20% and 25% of all local government revenues.  The formula for dividing the money is quite opaque, contentious, and clearly politically influenced.  This year’s allocation is more interesting than normal since several local governments are being upgraded to the status of direct municipality, and direct municipalities are accorded preferential treatment in the formula.  So my question is this: how much is it worth to become a direct municipality?

The figures in the media reports were splotchy and incomplete, and I wanted to know about how much everyone, not just the direct municipalities will get, so I started trying to track down figures from various government sources.  Here’s where everything bogged down.  None of the figures quite match up with the figures announced in the last few days.  The media’s numbers (which came from the Minister of Finance) are quite a bit bigger than the numbers from the Ministry of Finance website.  In fact, they don’t even list this pool as a separated category in the various tables of local government revenues, so I’m not sure if it counts as a subsidy or as tax revenues.  I have no doubt that all these numbers are right, in some way or another; I just can’t quite put them together.

I also can’t quite figure out the formula for allocating funds.  I found the law, in both Chinese and English versions, but it would take me somewhat longer than a few hours to figure out what exactly all those terms mean.  The main ingredients to the pool are portions of the income tax, the (business) operating tax, the commodities tax, and the land value increment tax.  In 2008, the predicted structure was as follows (I give the estimates rather than the actual figures because the world economy collapsed in 2008, and I think the predicted structure is probably a better window on the “normal” structure of the pool):

Billion NT
Income tax 82.3
Commodities tax 14.3
Operating tax 106.9
Land value increment tax 5.3
Total 208.8
Source: MOF website

(The exchange rate is roughly 32:1, so NT 208.8 billion is roughly USD 6.5 billion.)  The formula for allocating funds to individual local governments is very hard to decipher.  The main ingredients are the level of the government, population, the amount of tax collected in that area, land area, outlays for national health insurance, eye of newt, tail of lizard, and paprika.  So my version is going to be a bit wrong.  In other words, I’m not going to be able to answer the big question definitively.  But this is still informative.

Let’s look at numbers downloaded from the Ministry of Finance first.  They have data for 2005-2009, but for simplicity, I’m just presenting 2007 and 2008, since the only big change during that five year period occurred then.  In 2008, Taipei County was upgraded to the status of quasi direct municipality.  I don’t know exactly what this means.  I guess that Taipei County was classified as a direct municipality for at least some of the calculations, but, since they will have another huge leap in 2010, perhaps they weren’t accorded all the privileges of direct municipalities in 2008.

total total Tax pool Tax pool
population revenues revenues Allocation Allocation
2008 2007 2008 2007 2008
Billion NT Billion NT Billion NT Billion NT
Total 23037031 841.4 897.1 193.2 196.2
Taipei City 2622923 161.8 151.5 60.8 41.9
Kaohsiung City 1525642 64.2 59.4 22.9 12.4
Taipei County 3833730 73.2 91.1 10.4 32.7
Ilan County 460902 14.4 16.7 2.6 2.9
Taoyuan County 1958686 36.9 44.9 5.5 6.2
Hsinchu County 503273 16.1 17.9 2.7 3.1
Miaoli County 560397 16.5 21.0 3.3 3.5
Taichung County 1557944 33.8 40.9 6.4 7.2
Changhua County 1312935 29.0 29.5 6.3 6.9
Nantou County 531753 21.6 22.0 3.8 4.3
Yunlin County 723674 21.7 23.4 4.6 5.1
Chiayi County 548731 20.7 23.1 3.9 4.2
Tainan County 1104552 29.1 30.4 5.2 5.9
Kaohsiung County 1243412 30.7 31.0 5.5 6.4
Pingdong County 884838 25.7 31.1 5.4 5.9
Taidong County 231849 10.7 11.9 2.7 3.0
Hualian County 341433 14.8 15.3 3.0 3.3
Penghu County 93308 6.8 7.3 1.5 1.6
Jilong City 388979 14.1 14.2 2.7 3.0
Hsinchu City 405371 13.9 14.5 2.0 2.5
Taichung City 1066128 32.2 34.2 3.0 3.3
Chiayi City 273793 8.8 10.2 1.5 1.7
Tainan City 768453 20.3 23.3 3.0 3.3
Jinmen County 84570 9.2 10.2 1.1 1.2
Lianjiang County 9755 2.6 2.4 .3 .3
All townships 15985742 112.7 119.7 23.3 24.3

Townships are the level below county and city.  Only counties have townships.  (Cities have districts.  Districts do not have independent local governments and do not get any money from this pool.)  So if you live in a county, both your county and township government get a slice of the cake.  However, the township slice is noticeably smaller.  (There are 319 townships varying in size from less than 1,000 to more than 500,000 people.)

I purposely separated out Taipei City, Kaohsiung City, and Taipei County on this table.  The former two are direct municipalities, and Taipei County was made a quasi-direct municipality for the purpose of tax pool allocation in 2008.  Note how much bigger Taipei City’s budget is than everyone else’s.  The Taipei City government is much, much richer than any other local government.  In 2007, its total revenues (161.8) were more than double those of Taipei County (73.2), even though Taipei County’s population is significantly larger.  (This is not exactly a fair comparison, since the Taipei City government has more responsibilities than the Taipei County government.  Even so, the disparity is stunning.)   Kaohsiung City also did quite well in 2007.  Its total revenues (64.2) were quite a bit higher than either Taoyuan County’s (36.9) or Taichung County’s (33.8), even though both have larger populations.

When you look at the allocation of the tax pool, the advantages of direct municipalities are even more stunning.  In 2007, Taipei City (60.8) got six times as much money as Taipei County (10.4).  Kaohsiung City (22.9) got about four times as much as Taoyuan County (5.5) or Taichung County (6.4).

In 2008, the new system eroded some, but not all, of the advantages for the two direct municipalities.  Taipei County more than tripled the amount it got from the tax pool (10.2 in 2007, 32.7 in 2008).  This was still less than Taipei City’s 41.9, but the gap was much smaller.  Note the commensurate changes in the governments’ total revenues.

Another way to look at this system is to look at per capita distributions.  This is the same table, but with per capita figures.

total total Tax pool Tax pool
population revenues revenues Allocation Allocation
Per capita Per capita Per capita Per capita
2008 2007 2008 2007 2008
Total 23037031 36523 38943 8387 8516
Taipei City 2622923 61685 57756 23181 15986
Kaohsiung City 1525642 42056 38952 14981 8117
Taipei County 3833730 19094 23752 2707 8539
Ilan County 460902 31165 36238 5652 6278
Taoyuan County 1958686 18844 22945 2784 3189
Hsinchu County 503273 31944 35509 5408 6191
Miaoli County 560397 29496 37417 5869 6314
Taichung County 1557944 21667 26234 4081 4596
Changhua County 1312935 22086 22485 4826 5240
Nantou County 531753 40617 41292 7060 8024
Yunlin County 723674 30026 32355 6288 7096
Chiayi County 548731 37669 42175 7081 7669
Tainan County 1104552 26307 27536 4735 5326
Kaohsiung County 1243412 24676 24923 4432 5132
Pingdong County 884838 29063 35108 6080 6717
Taidong County 231849 46201 51524 11451 12877
Hualian County 341433 43268 44845 8899 9772
Penghu County 93308 72682 78049 15799 17337
Jilong City 388979 36253 36430 7061 7689
Hsinchu City 405371 34372 35857 4912 6068
Taichung City 1066128 30163 32037 2846 3127
Chiayi City 273793 32152 37318 5596 6205
Tainan City 768453 26438 30351 3854 4236
Jinmen County 84570 109285 120920 12590 14387
Lianjiang County 9755 267657 248484 26573 30310
All townships 15985742 7050 7488 1460 1520

If you ignore the direct municipalities, the system is basically progressive.  The more populous, more urban governments get a lot less money per person than the more rural, less prosperous, and more sparsely populated governments.  So Taichung City only got 2846 NT per person in 2007, while Taidong County got 11451 per person.  The three highest numbers belong to the offshore islands: Penghu, Jinmen, and Lianjiang.  This is reinforced by the fact that Taichung City, as a city, does not get any of the money for townships, while people in Taidong County, for example, got 11451 per person for the county government plus their cut (which was presumably higher than the average of 1460 per person) of the funds for township governments.

Of course, you can’t ignore the direct municipalities.  When you look at a per capita basis, the unfairness of the old system glares at you.  Taipei City, the richest area in Taiwan, got nearly as much per person as Lianjiang County in 2007, the poorest, smallest, and most remote place in Taiwan (or not in Taiwan, as the case may be).   Even after the reform in 2008, Taipei City (15986) still got five times as much per capita as Taichung City (3127).

Ok, what about the jump between 2009 and 2010?

Population (2008) 2009 allocation 2009

per capita

2010 allocation 2010

per capita

Billion NT NT Billion NT NT
total 23037031 360.2 15636 458.4 19898
Taipei 2622923 56.3 21465 58.4 22265
Xinbei 3833730 34.2 8921 60.3 15729
Taichung 2624072 25.8 9832 35.6 13567
Tainan 1873005 24.6 13134 33.0 17619
Kaoshiung 2769054 40.1 14481 54.4 19646
all others 9314247 179.2 19239 216.7 23265

Keep in mind that the figures reported in the media do not match with the ones listed above (360.2 billion is much larger than the 196.2 billion listed above for total spending, and 2008 and 2009 were almost identical), so I’m missing something in their accounting.  However, the relative sizes still match.

Note that the whole pool is growing by roughly 20%.  I’m not sure if the MOF is putting more of the same sources into the pot, or if it is putting revenues from new sources into the pot.  At any rate, the pot is bigger, and the MOF promised that no one would see their allocation decrease next year.  Overall, this table looks a lot less unfair than the previous ones.  In 2009, Taipei City’s per capita allocation was only about 40%, not double) more than the national average.  Taipei County’s per capita allocation is about half the national average, whereas above it was roughly equal to the national average.  So maybe I’ve been overstating things.

The big headline over the past few days is that Xinbei City’s allocation will nearly double in 2010, and it will surpass Taipei City’s.  The other theme has been politicians from Taichung and Tainan complaining that they are not getting their fair share.  In fact, Tainan and Kaohsiung seem to do ok in this system.  Taichung is the city that should complain.  Taipei City is still a winner, though not by nearly as much.  Even though it has roughly the same population as Taichung and Kaohsiung and a lot less land area than either, it will still get more money than either of those two cities.

Perhaps the most fascinating thing about this table is the last line.  All those other counties that didn’t get upgraded to direct municipality look like they will take a fairly big cut of the money.  Perhaps the new formula has taken away the advantage that direct municipalities used to have.  I really want to see what Taoyuan County gets in the new system, since it is the most similar county to the new direct municipalities.  I would bet that Taoyuan will get somewhat less per capita than either Xinbei or Taichung, though the gap can’t be too great.

Even though I can’t reconcile the two data sources, it is clear to me that the system has become much more equitable over the past five years.  It is no longer clear that direct municipalities get a grossly unfair cut of the cake.