2017年6月22日星期四

香港的寫字樓與零售樓宇分佈是否恰當?

載於信報2017年6月22日

近期中環美利道商業地皮創出歷史新高之總價(233億)及樓面地價(每呎5萬元),不難令人懷疑這是否供求失衡的結果,筆者借此機會在本文中探討商業樓宇之供應情況。
在過去三十多年中,本港寫字樓及零售物業的存量都在穩定增長;然而,寫字樓的增幅卻一直快於零售物業(【圖一】紅條)兩者的總樓面面積比例由1981年之0.62倍大升68%20161.04倍。

圖一:寫字樓面積增長持續快於零售樓宇


此次中環新地王的產生,可歸功於多項因素,包括現時的超低利率環境,及以香港作跳板拓展海外市場而日趨強勁的內地企業需求,令本港租金和寫字樓價格屢創新高。
然而,以上理由亦同樣適用於零售樓宇不斷增加的內地旅客豈不是也在對零售物業產生巨大的需求爲何寫字樓卻能在供應上作出更大的回應,大幅拋離零售樓面增長


消費模式轉變:從商品到服務


筆者認爲,寫字樓存量增速高於零售空間或與香港經濟結構的變化有關——越來越多的消費行為轉為在服務領域進行,而非單純局限與商品交易這層面上。對比三十多年來的表現,服務業開支的增長大大高於和零售金額:兩者比例由80年代的0.9倍增至21世紀初的2.5倍(【圖二】橫軸)。同時,似乎寫字樓與零售樓宇的存量之比也有對應的增長(縱軸)。除了在過去四至五年中,服務對零售消費比例出現短暫的波動之外(或是因爲內地遊客於數年間激增所致),兩組比例幾乎呈完美的綫性關係。

圖二:服務消費比重上升帶動寫字樓存量跑贏零售物業 

若由私人消費開支角度來看,可見服務類別的消費比重由1981年的43%增加至現在的60%;而不少服務類的消費主要就是在寫字樓內生產製造,甚至同時在寫字樓內消費掉。在同一時期內,需要更多零售空間作支援的食品和消費物品所佔比重由58%下降至47%(見【表】)。

表:私人消費開支:服務類比重越見上升

上述服務消費增加的趨勢更詳細地於【圖三】中呈現出來。反過來看,用於統計通貨膨脹數據所用的住戶開支籃子成份同時也可見到零售貨品比重下降的跡象(零售物品包括食物、飲料、衣物和耐用物品,見【圖四】紫線)。

圖三:服務佔私人消費開支比重愈來愈大 
   
圖四:實體貨品在通脹籃子中比重在大幅下跌


從實體到虛擬、由線下至線上


近年另一個新趨勢則是網上零售的爆炸性增長【圖5】,這意味著零售行銷會有越來越多在寫字樓(如電子商務工作人員經手網上購物)和工業或倉儲樓宇(為網上購物的物流部分提供支援)進行,而不佔用傳統的零售樓面。此趨勢對未來寫字樓或零售空間的需求會有更加舉足輕重的影響。

圖五:上網購物的港人數目(百萬)大幅攀升 


為何寫字樓租金仍未高於零售業樓宇?


既然服務業的增長更為強勁,那為何寫字樓租金並無顯著跑贏零售樓宇租金【圖六】?一個較簡單的解釋可能是寫字樓供應量增長亦同時大於零售樓宇,故而削弱了寫字樓租金的上升優勢。由空間角度去解釋,因為寫字樓的樓面可以堆疊於多層大廈內而不影響其實用性,但是零售物業需要人流量大、交通方便、及醒目當眼的地段方為零售商青睬,因此多層商廈中的零售樓層一般都局限在最低幾樓(通常不過三),這在某種程度上使零售樓面較寫字樓稀缺,租金亦較後者為高。

圖六:過去廿年寫字樓租金跑贏零售,但在絕對值上仍有距離  


那麼未來的發展會是怎樣呢?寫字樓的供應量會繼續領先零售樓宇嗎?答案是未必。因為:一)隨著內地遊客數量回升並創新高,更多獨特主題的多層購物中心將會落成甚至由寫字樓宇改建而成;二)移動/共用辦公和無固定辦公桌的新辦公方式可以減少員工人均寫字樓面積(有企業預計轉向無固定辦公桌和共用空間能減低多於一成樓面需求),更遑論機器人的出現可能對職業白領帶來的威脅及造成的相應寫字樓需求量的減少。

無論未來怎樣發展,香港應能保持其區域總部及購物天堂的地位,而本港亦將繼續受惠於亞洲較高速的整體經濟增長。因此,本港寫字樓及零售樓面的長線表現都應該會優於其他國際城市之相應板塊。



筆者特別鳴謝香港中文大學學生賴子萱對此文的貢獻。

2017年6月16日星期五

Bureaucratic inertia distorts Hong Kong’s taxi market

 

The recent steady stream of reports on problems related to the Hong Kong taxi industry, from drivers’ refusal to accept rides to overcharging, from sky high taxi licence prices (record high of HK$7.2m back in June 2015) to the Government’s persecuting innovative hailing businesses such as Uber – all point to the worrying trend of an ever worsening business environment in whose excellence Hong Kong has traditionally prided itself.

This article analyses the deteriorating transport ecosystem in Hong Kong by looking from a supply and demand perspective, hoping that more appropriate policies can be adopted to put the increasingly suffocating travelling experience HK citizens face back on its former flexible, high-quality, and affordable track.

Govt limits on licence supply leads to sky high prices


The Hong Kong business environment of recent years has been characterised by worsening service standards, astronomical asset prices, and purging of new entrants or small players, resulting in a overwhelming domination by the establishment (aka big corporations) while the consuming public suffers. In the taxi industry, this started back in 1998 when, under the pretext of considerations for the common good of the taxi profession, the government completely ceased issuing further taxi operating licences. This move sowed the seeds of a major bull market in license prices, which has massively outstripped the increase in the income of the riding public (Chart 1).

Chart 1: starting at 100, income rose a mere 34% vs taxi prices up 286%

The fact that prices of a right to drive a means of public transport can soar to the value of a private home is nothing short of incredible, even though this phenomenon can be explained by the simple rules of supply and demand. It is clear from Chart 2 that from the 1970s to the mid-1980s, due to the rate of increase in taxi licences being higher than that of housing supply, the ratio of taxi licence price to home price index fell significantly to within a range of 0.5 to 0.8 (see red highlight in Chart 2). This is ample proof that when supply is abundant, a utility item will not turn into a commodity of speculation.

Chart 2: lower licence supply resulted in licence prices rising more against home prices

The situation reversed from late 80s and persisted throughout the 90s, when the increase in taxi licence numbers fell behind housing supply. As a result, the ratio between taxi and home prices surged back to the 1.1-1.5 region (see yellow highlight in Chart 2). This would have been the best time for the Government to increase the supply of taxi licences, but sadly, it unwisely ended new supply of licences from 1998 onwards, resulting in a total windfall for licence holders/speculators, when the licence price to home price ratio more than doubled to a high of 2.8 in 2005; although the ratio fell back after the global finance crisis, it still trades at an elevated range between 1.4 and 2.0 (see green highlight in Chart 2) – this is a level triple the norm back in the 70s and 80s, and formally sealing the licence’s status as an instrument of speculation instead of a document to authorise provision a service to the public.

Bureaucratic forever behind the curve


The role of government is best limited to providing a fair environment of exchange, so that the citizens can freely supply and consume products or services in an open market. Looking at demand for transport services, it is not difficult to conclude, from looking at the changes in population, housing supply, and private car numbers since the handover, that the city’s demand for taxi services remains severely unsatisfied.

Since the government stopped issuing taxi licences in 1998, Hong Kong’s population has risen by 12%, its housing stock by 29%, and private cars on the roads by 60% (Chart 3)! If these are taken as the manifestations of underlying demand, then should there not have been extra taxi licences of 2,176 (i.e. 18136 x 12%), 5,259 and 10,882 respectively, according to the three demand proxies? To take a more extreme example – the number of visitor arrivals which has surged 459% in the same time – the number of taxi license should have risen an even more astronomical magnitude. The negligence that has led to this level of disconnect between demand and supply is simply unforgivable.

Chart 3: taxi provision is severely lagging all demand indicators 

The classical supply-demand curve explanation to the phenomenon is simple – when supply is arbitrarily fixed (see orange line in Chart 4), continued rises in demand (blue lines) will push prices up (P0 to P1). The natural and appropriate way to adjust to changing demand situation is to allow the market to adjust supply levels such that both the numbers and the prices of taxi licences are determined by the free interaction between the buyers and suppliers, not by remote mandarin officers sitting behind closed doors.

Chart 4: inflexible supply pushes up licence prices, putting up barriers to entry


Govt policies promote rent seeking


Since Hong Kong’s handover, we have witnessed more and more reports of corporatist policies favouring monopolistic behaviour while penalising new and small market players. In the context of the taxi industry, the abrupt and continued suspension of licence issuance has resulted in supply shortage and surge in licence traded value – and the increased barrier of entry (due to higher capital requirements) means that more and more licences are now held in the hands of hoarders and speculators.

Under normal circumstances, owner-operators should possess one licence, while those holding three or more taxi licenses can be safely classified under corporate owners or investors. Since 2011, when statistics have become available, the proportion of corporate ownership – that is, holding more than 3 licenses – has gone from 18.4% to 22.7% in just 5 years. That is a cumulative growth of 4.3 percentage points (Chart 5). In contrast, the proportion of single licence holders fell by an even larger 5.4 percentage points. This suggests that taxi licences are being concentrated from the drivers (which make up a maximum of 52.6% of total now) to the well capitalised investors.

Chart 5: taxi licence ownership increasingly concentrated


Drivers are forced to take risks


When a humble licence to drive a means of public transport has morphed into a tool of investment for wealthy, we know that policy making in this area has become abused. Sadly, the Government has ignored the issues for many years, leaving taxi drivers and passengers to pay over-inflated rents and fares.

The shortage of taxi licenses results in increases in taxi rentals for drivers. In order to breakeven, the drivers developed revenue maximisation tactics such as hire refusal, taking longer routes, and overcharging. This is only one of the outcomes out of the problem of imbalance between supply and demand. This sad symptom of the policy error is clearly visible from the prosecution figures where the number of drivers charged for ride refusals rose nearly sixfold in the past five years (Chart 6) – the deterioration of service standards has become systemic.

Chart 6: prosecution for refusing hire were hugely increasing


If on the other hand, the number of taxi licences is allowed to fluctuate freely in response to market demand, then the licences will lose their investment value, and the rent charged on the drivers would decrease at the same time as it now merely needs to cover the operating costs of the vehicle. Just a one-third drop in taxi rents will allow the driver to earn the same level of income while passing on a 10% saving on fares to consumers (Table 1). If, however, the driver chooses to reduce his working hours by 10%, the riders can still save 2% on their fares compared to the status quo, is this not a win-win situation for all?

Table 1: Scenario showing how low licence fees lead to better economics for drivers and riders

Planned economy approach deprives our freedom of choice


Under the suffocating atmosphere of over-regulation, more and more people are turning to taking the MTR, as shown in Chart 7, gaining the dubious pleasure of daily sardine-packing and labyrinth walking in the tunnel system of our underground system. On top of that, the increasing failure rates at the railways leave the population with next to no choice when taxis become such a hostile form of public transport.

Chart 7:


The dysfunctional state of affairs caused by artificially restricted supply of taxis on our roads has led to the market share of taxi rides falling precipitously despite major surges in the length of roads, population, and tourist visitors. This trend is particularly pronounced since the government ended issuing new licences from 1998 onwards (Red arrow in Chart 7), is it any surprise that complaints about taxi services go sky high and new technologies such as ride hailing apps become all the rage?
When public criticism mounted after the bureaucrats tried to kill off Uber, their only bright idea was to placate us with a "luxury taxi", this clearly shows how the illogical mind-set still persists in the power corridors where policies tend only lead to duplicating, fragmenting, barrier-erecting, and oligopolistic market behaviour. This self deceiving mentality completely disregards the most common sense approach of giving the power of supply back to the market and lowering entry barriers.

In the meantime, current policy continues to condone the explosion of private cars on the roads, increasing congestion exponentially, when taxis are the most natural substitution to private car journeys. Such short sighted policy making is just another piece of evidence that the government has lost the plot.

With special thanks to Mr Chi Chung Kelvin Ng and Kwok Yan Chiu for their contribution to this article.

-----------------------------------------------------------------------------------------------------
Please follow us!
LinkedIn: https://www.linkedin.com/company/b&mm
Blog: http://yulun2012.blogspot.hk/

2017年6月6日星期二

政府預測:盲目樂觀抑或是故意誇大?

載於信報 201765

鑒於兩個月前建制派與泛民於立法會為9000餘件資金申請爭論不休,筆者藉此機會審視政府是如何在呈現其各項遠景時利用預測的可塑性來減少公衆反對及立法會的監查。

人口「增長」:一切基建的推手

爲了建設新的大型項目,當局首先需要向公衆闡明投資這些項目的必要性,而其中最重要的原因之一便是人口增長—越多的人口,就能更理直氣壯地建造越多的房屋、越多的公路、越多的設施、等等。

圖一: 人口推算總是過於樂觀


因此,人口推算數據頻頻成為被政府誇大的對象,似乎是理所當然的。由【圖一】可見,回歸後幾乎每輪推算都會在發佈後短短的一至兩年間就被下調。也許這些常常差誤甚大的人口推算正就是所有近年基建項目資金審批的推動力所在。


19972015年的八次人口推算中,僅有兩次被上調過,儘管如此,這兩個推算還是最終雙雙被再被調低。此外,該兩次上調的時機亦可圈可點:第一次上調正撞著2004年沙士疫情結束,政府亟需大興土木來沖喜疲弱的本土經濟;而第二次則適逢2007年特首換届,短暫吹大人口推算絕對令新政府更水源充足,方便其在各領域大展拳腳。


預測越長期,偏差越加大

爲了量化每一年最初推算之人口與最新推算的偏差,筆者做了一個簡單的分析,發現官方推算誤差會逐年擴大【圖二】。例如,1997年時初次推算之2016年人口為821萬,但是等到2015年最新推算時已縮減至735萬,兩者差幅達到10%。若計2029年人口推算,則新舊推算之誤差幅度更已增加至最高峰的12%。


圖二如果沒有最新推算下人口收縮這現象出現,新舊推算誤差還會更趨膨脹 



由另一角度去看,每次人口推算都會被後來的新推算推翻【即需下調,見圖三】,可見政府在制定此一基本統計數據時是有一個慣性的樂觀傾向。

圖三所有人口推算都有樂觀傾向

以上往績顯示政府的人口推算幾乎是一邊倒地樂觀:實數或最新推算數在短短數年間會較原始預期低出達12%之多。


唯一例外。。。

較為樂觀的人口推算可以歸根於官僚們試圖為他們的各項計劃預留一些回轉餘地。畢竟,不會有人因爲路上車流順稀疏而去指責政府,但是若果剛開通的新路由於預算不足而終日塞車,則主事官員必會招來眾多詬病。

 若以負面角度來解釋這種慣性樂觀的人口推算現象,不難會理解為:這些人民的公僕故意用較高人口推算誤導大衆以獲取更多資源去從事其眾多取自公帑的計劃,并且當項目逾期或超支時可以不須負責。 

這種懷疑態度甚至可以解釋爲何突然間政府在2015年所作之人口推算能打倒昨日的我,竟然預測2039年之後的本港人口將大量減少(圖一棕色部分),這與之前慣性的人口膨脹假設無疑是南轅北轍。此矛盾在看清背後動機後則不難理解矣:自從2015年開始,政府就積極籌備成立「未來基金」。即是,政府要說服市民在納稅賦餉以外額外多掏腰包,去供奉另一個由官僚把持,市民卻無從決策的基金。爲了説服大眾這隻基金不可或缺,最簡單的方法就是向市民展示這人口下降情景下之慘情。如果蟻民不乖乖掏出自己的血汗錢交由聖賢般的父母官打理,他們將會一貧如洗。可見,這突如其來的人口驟降推算無非是用一種看似科學的、難以反駁的論證來説服港人認同「未來基金」的設立。在這種解讀下,政府的慣性樂觀能在瞬間變為世界末日般的悲觀,就不難理解了。


預測失準的案例分析

在現實生活中,隨手可以找到政府(或其天價聘請的顧問)偏離現實預測之個案。

1) 形象工程一:港珠澳大橋  
「形象工程」是一個具中國特色的詞語,意謂那些主要功能不是為滿足人民需要,而是爲了滿足虛榮或是幫助幹部晉升而進行的工程項目。近年,內地官員的仕途在很大程度上取決於在GDP增長數字上達標,而衹要以人民的名義多澆水泥,GDP增長就不難超額完成。 

港珠澳大橋似乎正就是這樣的一項形象工程,原因如下:  
1) 儘管投入1100億港元(佔年度GDP4.5%)的天量資金,該工程竟然不包括鐵路客運;如要打通珠三角兩岸經脈,鐵路運輸應是任何基建的首選模式。

2) 最初大橋預計落成開通后需向每車次收取140港元車資方能於30年内回本。然而,其建造成本在不斷攀升的同時,新的車流推算卻在不斷減少,導致回本車資已飆升至380港元/輛,近乎原來的三倍!不難想像,未來的造價水平極可能繼續攀升,而車流量推算卻繼續降低,令平均回本車資更不經濟。 

3) 由於缺少泊車設施,道路容量問題以及邊境檢查的複雜性,當局甚至在該項目完成前就下調了跨境車流量的預計。

圖四短短十年內,港珠澳大橋車流量預測已回調高達三分之一
  

如【圖四】所示,當最新推算在2008年發佈時,大橋的預計車流量相較2000年初始預測下跌38%,即多於三分之一!而期間約一成的人口推算誤差都不足以解釋當年過度樂觀的預期數字。唯一能解釋此一龐大誤差的原因,必是當局有心誇大成本效益以博取通過立法會審議,或減少被削資的風險。 

2) 形象工程二:高鐵項目 
廣深港高速鐵路項目已經出名地超支超時:總成本已由當初的650億港元飆至840億港元,漲幅達29%(政府已因政治原因為此造價封頂,以後再多出部分雖由港鐵承擔,實際上仍是要納稅人埋單),落成時間也由2015年8月推遲至2017年8月,令施工時間延長約20%。因為項目占據了九龍最中心最優越的地段,因此庫房不但損失天文數字的土地收益,而且高鐵所帶來的時間優勢也被逐屆政府消磨殆盡:1998年時通往廣州的48分鐘車程,現在已經拉長至超過一小時(原因包括四站變六站,使一條跨境高鐵變成通勤慢車),而到北京的車程也由七小時變成了九小時。相較于現有一小時五十四分的直通車旅程,以及衆多可由香港18區出發、沿途可於居民區方便停靠的跨境巴士,需要在遠離市中心的郊區落車之高鐵就顯得畫蛇添足。 

也許當局想為成本超支及工時延誤遮醜,短期內的乘客量預測被上調(在現金流折現計算上可以增加項目的净現值Net Present Value),然而遠期乘客量的推算仍被調低6%,如【圖五】所示。

圖五:高鐵客流量預測誤差

 


 3) 立項前預測多會過度樂觀 

上述討論證明在一般情況下最終實數往往較最初預期縮水。然而,當項目尚未去馬前,隨後之預測數字卻會較最初預期膨脹,特別是落成目標日期尚在遙遠的未來時。這種逆常態的現象不難解釋:只有如此改動數據,方能令當局更便利地通過審批其發展大計。

此種行為可在現正籌備中的東涌新市鎮發展計劃見到。爲了盡量獲取更多的公共資源,政府在2015年做的2023年人口推算比2007年的推算上調了19%之多【圖六】。


圖六:東涌新市鎮人口預測被大幅上調
 

這次上調動作可以歸因于2015年立法會對東涌新鎮計劃第二階段可行性的辯論:政府如欲在立法會獲批通過此一龐大工程之撥款,則必須呈現一個落成後之美好願景,而越吹越大的需求,則為辯論提供可靠的「依據」。 

4) 爲何人口推算忽然變超級悲觀? 

讀者應記得【圖一】中,政府突然從2039年開始推算香港人口將會下跌,一反多年來不斷上調人口推算之習慣。

到底是何事導致如此出乎意料的一個一百八十度轉變?原來,從2015年12月開始,政府就開始對人口老化問題表示關注,以及如何為將來供養大量老人而作準備。

當然,這個憂慮的首個謬誤就在於假設市民在沒有慷慨的父母官照顧下就無法照顧好自己;其次,爲何在如此巧妙的時機,政府竟能果斷而全面地推翻自己長期以來所堅守的人口必增信條?如【圖七】所示,東京這個可能是地球上人口最老的城市,在過去20年間人口數量亦持續增長。光是這個例子,就足以令人質疑爲何香港的人口數量會下降?作爲一個更加開放外向的主要國際城市,香港不是更較東京能吸引人才流入嗎?

更有甚者,一個能不斷增長的城市,必能輸入人手及技能去照顧自己的耆英階層而無須依靠另一名目的稅項——即「未來基金」是也。

圖七:東京人口隨著日本年齡中位數一起攀升 
 

官僚所忽視的一個更重要的趨勢就是,更長的壽命賦予長者在傳統退休年齡後繼續工作的能力:與其在55歲退休後等待慈善救濟,他們可以輕易地在60甚至70多歲之齡仍然高率地工作,繼續貢獻他們寶貴的經驗及專長。

事實上,在東京及本港,老人的失業率似乎都比整體人口的失業率低【圖八】,兼且,香港長者的勞動參與率在不斷上升【圖九】:如果香港人口年齡分佈像東京一樣,則長者勞動參與率(已經從2010年之27%上升到2016年之33%)將進一步向39%邁進。
  
圖八老年失業率遠低於社會平均水平
 
   
圖九:老年人勞動參與率正在提升
 

還有,長者一輩應是社會上最富有的年齡群(香港缺乏相關統計,美國數據見【圖十一】),而彼等更絕不是收入最低一族【圖十】。必須注意,較低的收入可能正正反映長者不想一直工作,而是更多休閑活動的現實,且其收入更受退休儲蓄補貼。
  
圖十:按年齡分組的香港月收入數字
 

圖十一:各年齡組的財富分配狀況:老而彌富!
 


設立未來基金附有巨大的道德風險:一旦政府將退休保障强制化,市民便會停止部署自己的退休計劃,轉而依賴政府的施捨。這就是導致歐洲主權債務危機的起因;經過數十年高賦稅與高福利環境的「滋養」,歐洲自力更生的意識已被這種依賴性文化取代。與其將長者視爲一項負擔,倒不如當其為一項寶貴資產;將退休保障集中於官僚手中無異於奉獻自己的經濟自由,甘願成為政府福利的奴隸。

上述分析清楚顯示,隨著壽命的提升,60歲不再一定是職業生涯的終點。與其徵收苛捐雜稅使人民喪失照顧自己的能力,我們更應該鼓勵自食其力及以家庭為本的倫理觀念,因爲這個二十世紀方才出現的福利信條已為自古以來傳統的社會構造造成足夠的傷害矣。

結論 

上述例證足以證明官方預測非但未臻完美,有時甚至連客觀都做不到。官僚常會為自己的目的而扭曲預測。公僕的各位主人必須對每一分以我們的名義而花的錢保持警惕。而保護自己錢包的第一步就是挑戰每一個官方預測的合理性,無論其表面動機是如何的良好。


筆者特別鳴謝香港理工大學學生沈振翾協助收集及整理本文相關的數據及圖表