我和兩位 blog 主人不同的地方是, 我有幸抽到一個在北投的「爽單位」, 過著當兵的人都會說「爽」的軍旅生涯, 這樣會造成我對軍隊這個組織, 有著與兩位 blog 主人不同的評價嗎？

恰巧完全相反, 當完兵後, 我不僅僅支持他們的立場, 甚至有過之而無不及。

軍隊這個組織麻煩的地方是, ratchet effect 的問題由於組織設定所賦予上級對下級的絕對權威, 導致上下在任務溝通和合作上, 有著嚴重的 coordination failure, 更使得上級的目標和給予的資源, 在成效上都大打折扣。有的時候我甚至想, 在我的小單位裡, 中隊長官們都稱不上「無腦」, 所有在軍中會出現的問題都還是出現了, 那基層單位豈不是更加悲慘？他們要資源沒資源, 要預算沒預算, 最多只有多了一些的主官裁量權, 那這樣怎樣才能持續提供良好的國防公共財？

在不揭露國家機密的狀況下, 討論一下這個問題好了。敝單位是標準的幕僚單位, 目的是定時把蒐集來的資料呈上去, 裡面也包括也寫「研析」, 意指要分析新聞事件以供上級參考, 但由於我們的再上級單位, 擁有督導敝單位的權責, 更由於敝單位的成品, 閱讀對象是包括副部長在內的高級軍官和更上級單位的長官, 以致於我們的成效就是他們的成效, 這種誘因體系理應讓上級長官與我們站在同一線, 不過可惜的是, 也由於更上級長官對下面的掌握可能性趨近於零, 使得我們的上級指導長官們, 只求要看到「高高層」看了滿意, 更加雪上加霜的是, 這些上級指導長官熱愛「體察上意」, 只要做出來的東西與他們的想法不一致, 動則威脅中隊長官和弟兄的福利, 也並未有實質的指導和協助, 但在接受局長表揚時, 擺出的態度卻是「是因為我們的指導才有今日的成效」。這僅是軍隊體系內 incentive scheme 失誤的例子之一, 更是「不合理的要求是磨練」這句軍中「傳統智慧」的反例。

感想文就要以感想文的寫法做結。還記得端午節愛民打掃的時候, 附近可愛的文化國小小朋友看到我們過去幫忙的弟兄時, 天真可愛的說：「阿兵哥耶！」, 我慶幸這些小朋友以後不用經歷這種過程, 也願他們到了這個年紀的時候, 政府政策支票可以不如預期般的兌現。

男中音是威爾斯男中音特非爾 (Bryn Terfel)

在演出兩次以後, 指揮 Sir Roger Norrington 講了這段有趣的話：

I overcome with patriotic pride. How about you, Bryn? No? I think we better do a bit more than that … One more, in Gaelic.

這個舉動換來全場的笑聲。

雖然可能有可能是我過度解釋, 但是在台灣聽到這段話, 心中的感覺是非常複雜的。特非爾是威爾斯人, Rule Britannia 是標準的大英帝國愛國歌曲, 對一個威爾斯人來說, 雖然可能不如蘇格蘭人感覺強烈, 但畢竟總是「大英格蘭主義」的產物, 而對這些「被併吞的人民」來說, 代表感受又是不一樣的。

今天把英格蘭愛國歌曲改為用加爾特語唱, 或許是希望藉由這個機會, 讓各地不同文化, 但是同在「大英帝國」統治底下的人民, 用不同的方法, 去面對過去的歷史。同樣的事情如果發生在台灣, 大概只會落得一個「操弄族群」之類的名號不了了之, 之前民進黨執政時代演出蕭泰然時, 也似乎也落得同樣的名號。

當然, 用台語唱「中華民國頌」, 或者用中文唱「台灣翠青」, 絕對不是和解的本意, 在我認為, 族群問題不是不存在的, 雖然解決這種問題需要時間, 但絕對不是不去提他, 當他不存在, 這個問題就會消失。

歷史的悲劇應該被記得, 但是強調原諒, 有的時候又是太強烈的存在, 如何和解, 那又變成一個難以回答, 但是必須重視的問題。

或許英國的族群問題沒有那麼簡單, 但是如果某一天, 台灣可以向英國人這樣, 在公開場合承認過去的歷史, 並用自己的方法給予尊重的時候, 族群在台灣才不會變成一個爭議性的議題, 而我們才能真正面對歷史。

後記：

寫完這段文字, 怎麼覺得自己像在寫中天政論節目陳女士在水果日報的專欄 ……

記者套話的技巧很差，他才問我兩個問題，我就知道他想要的答案。天生反骨，我越知道他要什麼回答，我越不想說。 我的回答印象中如下：

（地點：自由橋的單車道，正在等紅綠燈）

T台記者：請問你常常來這裡騎腳踏車嗎？

我：沒有，第二次騎！

T：你不覺得，愛河這段有些味道？

（心想：你都有答案了，還要問我？）

我：還好耶！已經比以前改善很多。

T：你覺得還好，那愛河哪一段的味道比較需要改善？

我：我都從上游騎到五福路那裡。我覺得愛河整治的不錯，下游整治的很好，上游也不錯。

（綠燈！）

T：所以你的意思是，愛河這段不需要改善？那麼哪段的味道需要改善？

（我猜他是故意會錯意）

我：不是，我是說，這段確實比較需要改進（或比較有味道）。

T：你覺得原因是什麼？

（紅燈！心痛！）

我：可能是家庭廢水吧！我也不知道，我不是專家

T：你覺得還有哪裡需要改進的嗎？

我：我覺得比起愛河的味道，腳踏車道需要等紅綠燈比較需要改進！

（以下連錄都沒錄）

我：淡水河也很臭啊！特別是三腳渡的那一段

T：我們也有做淡水河的

（也不曉得是真是假）

我看到T台記者有跟上次徵才博覽會的自由時報記者得不到我說「景氣不好，找工作不容易」的答案時一樣落寞的眼神，想說應該跟上次一樣不會被報出來了吧！沒想到TVBS這麼會剪接！（還是那時候已經五點了，記者要下班，不想再等另一個單車騎士？）學姐說，他以為我也覺得愛河很臭，可是我解釋之後，他也覺得我其實講得很vague。看樣子，T台斷章取義的技術不錯！

台灣的記者做專題，多採速食主義，沒有深入了解，先預設立場，再套民眾的話，就像先射箭再畫靶，以為這就是證據，這樣操縱輿論，無助於反映真實社會現象。 對我而言，愛河已經夠不臭了！因為我一向在與汽車共騎時，會戴上頭巾當口罩。但是一到河濱，我就會把頭巾拿掉。在記者「提醒」愛河很臭之前，我完全沒有感覺。反倒是，一些愛河單車道的設計讓我第一次騎就深深地感到設計不良：

1. 單車道要等紅綠燈

2. 七賢路（應該是）那裡有一個高1.9米的涵洞要穿過，在腳踏車上仍低於1.9米的人應該很少喔！

3. 有很多段落做木棧道，木棧道當單車道比較不平且損耗快。

4. 在與中華路交叉口附近，有一組單車停車處居然是蓋在高起來的防撞磚上面，一般人應該不會扛單車上去停吧，這種單車停車處使用率低自不在話下！

信手拈來，就有這麼多設計不良之處，比起愛河的味道更令人感到不舒服，記者卻未真實反映騎車民眾的意見，只反映了不了解愛河單車道的T台老闆的意識形態！

I am not a macroeconomist and done poorly in macroeconomics and finance. However, I do remember the cost of government substitution to civil purchasing power inevitably exits, and stimulus plans are by no mean to excuse from these effects of pushing up price level, or even worse, causing inflation due to the nature of public spending. The cost of “reviving” from recession is to accept a higher price level, by the Keynesian story.Interestingly, in the debate of public policy, it seems this kind of effect is totally ignored. At least it is the case in Taiwan.

Since I am not an expert at these fields, I planned not to discuss the current problem in the manners of macroeconomics and finance. To my limit knowledge of the ongoing economic crisis, the origin of the crisis is partly because of the evaluation of performance of financial firms. The performances, and the bonus, composing the most part of their income, of executives are determined by the company’s short term profit, or few years at most, and they would get compensated if they are fired due to the performance, or golden parachute in Wall-Street sense. The incentives scheme is to create short-term profit and disregard the long-run effects to the shareholders. Financial institutions, holding many assets from the public, will cause much more severe damages than other kind of firms. If the major financial firms take the same strategy on performance evaluation, aggregately there would be too much resources allocated at the assets with too much risk. Clearly, it is a principle-agent problem.

If we agreed the financial crisis is a principle-agent problem, the financial supervision is a solution to the problem. However, another problem arises: to the limited knowledge we have, how can we measure it is professional or self-interest decisions made by the executive. This is the essential problem of regulation: the information problem. Moreover, in the cases of stimulus plans, it is likely to empower the administrations with plenty of resources and less restrictions to fight recession due to the time constraints. The term of a government is merely measured as “long term.” The time constraint caused by elections forces a democracy government to achieve the results of “fighting recessions” as fast and obvious as they can. Long-run effects are often, and almost always, forgotten. For politicians, the government “dies” in less than ten years anyway. Social optimality conflicts with politicians’ goals. Thus, the paradox of government policies is: The solution proposed to solve a principle-agent problem causes another principle-agent problem with a larger scale!

I am not implying government has completely nothing to do to cope with the economic problems. As I mentioned before, some failures of coordination can be solved by the move of government policy. Nevertheless, it shall be more skeptical to government policy due to the purpose of avoiding damages by the inappropriate measures out of the kindness to relieve the suffering from economic recession. Standing on the other side of majority, I shall say: “Sorry, but I am not a Keynesian, because long run is not as long as we, especially politicians, imagined, and we would have not dead or be dying then.”

Buying American Export will kill 2 birds with one stone. Since many savers are developing countries where there are many opportunities to gain productivity (and living standard) by investing and spending. The productivity gain alone can easily pay for the initial investment, result in few or no debt in the long run. This is lest true for developed countries that are already sitting on the technological frontier, and are thus at the inefficient end of diminishing return scale.

Buying American, will mean that the American government wont have to embrace the Keynesian option of trying to raise even more debt to spend their way of a recession, again. It will selectively boost America’s most competitive industry while reducing American debt and spread confident all around.

Why is there no sign of this happening? A lot of this has to do with the other rational for the build up of foreign reserves by the savers. Namely, to accumulate enough dollar to counter any excessive devaluation of their currency. In current crisis, the savers are even less likely to spend their savings, without which the savers would be vulnerable to currency fluctuation.

End game. In order for the savers to be comfortable enough to spend their savings, two things need to happen. (1) a coordinated spending plan with the participation of all major dollar holders and (2) a guarantee to bail out in case of excessive currency devaluation by stronger and more independent IMF would be needed.

related stuff to read.
On liquidity crunch – 『Baby-Sitting the Economy.』
On 『The Causes and Consequences of the Economic Collapse『

UPDATE 090403: G-20 Leaders To Give $1 Trillion To IMF, World Bank

G-20 participants announced a tripling of loans available to the International Monetary Fund, to $750 billion, a $250 billion expansion in a special IMF fund to help members\’ foreign exchange reserves, and $250 billion to the IMF to support trade. They also agreed to sell IMF-held gold to poor countries.

Thats an okey amount of money, but the point is that they prove that they are willing to use IMF and WB and they are willing to put resources into it. However whether this will be enough to get the savers to start spending will depend a lot on how exactly this money will be use, and who will have power over it. I personally knows too little about the inner working of the IMF and WB to offer any insight.

UPDATE 090405: World Wad
A world currency is another way to eliminate currency shocks, naturally. SDR, if its rely upon more, it has a change to become a predecessor to a world currency, just like ECU is to Euro Lets hope so. .

The proposition is awkwardly framed, i mean come’on, arguing for the proposition is like arguing that entire political and economic elites agrees over the remedy needed to recover from the current financial crisis, which is an impossible position to defend. This forced DeLong to declare defeat in the opening sentence of his opening remarks. And belatedly, moderator re-frame the debate to be 『whether we—or rather, economists and policymakers—should be [Keynesian].』 on the third day.

anyway, here is the problem

There are two recession dynamics at work: (1) plummeting real estate value and consumer spending is discouraging or disabling mortgage payment, leading to piling bad debts that is spreading through the financial sector via securitization and derivatives, banks thus concentrating their capital on deleveraging and not lending to business for expansion. (2) lowering consumer spending is reacted with output contraction and increase unemployment, that further reduces spending and thus the deflationary spiral ensues.

and the Keynesian Proposition

Delong is obviously not against government intervention to get the banks to start landing again or using every possible monetary policy to stimulate the economy. What he is saying that those measures have limits. Monetary tools is already exhausted. and not without politically unacceptable amount of taxpayer’s money can government keep the banks capitalized.

Even with recapitalized banks, in the short-run, under overwhelming uncertainty and pessimism, healthy business that would have invested, will not; innovations that would been commercialize with venture capitalist‘s funding, will not; employed people would have leave their job to start a new spin-off, will not. And there are no guarantee that the money from the recapitalized banks wont just be used to fill another bubble, which is precisely what happen when federal bank pump money in response to the busting of Dot-com bubble in 2000 only to walk straight into a real estate bubble. Therefore, DeLong argues that in the absent of private investment and deteriorating consumer spending , government who can borrow cheaply has to step in to invest and spend. DeLong also believes that government investment in public goods like infrastructure, basic research, education, broadband access in rural area can create new innovation and business opportunities that fuels economic growth in the future (off setting tax increase needed to pay government debt), and fights deflation now.

the criticism

The fear of fiscal stimulus is that it is hard to execute effectively, because government (by design is slow), wont move quick enough to start projects to deal with short-term problems. By the time government is able to initiate most of its spending, some private business that are ready to invest, will be force to compete with the government for capital (goods, service and labor). Even worst, if in a state of panic, the government is legitimized to pursuit projects that does not have benefit worthy of its cost in order to keep people employed, people will be under an illusion of job security, and not looking for new job, retrain, or staring new entrepreneurial enterprises.

The other problem of allowing government to allocate resource is that it is often inefficient and it expands politician’s power and attract lobbyist, which by the way is already happening. Obama promised greater transparency to mitigate this problem, but even some Democrat supporter are skeptical.

Furthermore unlike isolated national economic recession, this is a global recession where US government borrowing will have an adverse effect of draining the fund other economies needs for their recovery, subsequently weakens demand for US exports that is crucial for recovery. Unfortunately, global dimension of the recession is noticeably absent from the debate.

to conclude.

If the fear of massive unemployment is that it will lead to social unrest and deflationary spiral, then the government should simply increase and extent unemployment benefit. When people’s livelihood are secure and not occupied by bogus project they will have time to fund jobs, retrain or invent the next growth inducing innovation. Government spending in public goods can be very efficient considering the low cost of everything in a recession. However as Zingales points out these these projects 『should be argued on their own merits, not as a stimulus.

So, am i thinking Keynesian? I am not too sure, i mean the label Keynesian is so loaded with different meaning. I am however definitely for unemployment benefit in a recession.

這陣子老闆幫經建會做了不少救經濟的研究，手頭上正好有一些從經濟部下載的資料，雖然我不會老闆在威斯康辛學的艱深統計方法，也不會跑技術性的統計軟體，不過我聽過三重徐先生介紹的貿易重力模型(Gravity Model of Trade)。趁著放榜學校都沒有錄取的空檔，用STATA跑了一些圖，看看資料和貿易重力模型告訴我們什麼故事。

貿易重力模型和牛頓的law of gravity邏輯相似，簡單說，就是兩國的貿易流量與兩國的經濟體大小成正比、與距離成反比。我創造變數ydist為（貿易對象的GDP/貿易對象與台灣的距離），放在X軸。Y軸則用貿易對象與台灣進出口總額，畫簡單的scatter plot。資料內容以經濟部篩選的中華民國主要貿易夥伴為主，期間為1992到2007年。另外，值得一提的是，圖中的紅線為以該年度的最佳配適值。

先看1999年(陳水扁執政前一年)：

中國與台灣的貿易量顯然低於模型的預測值，告訴我們1999年，台灣與中國間確實有貿易管制存在。

再看2005年：

中國與台灣的貿易量已經顯著地高於模型的預測值，這個事實否證了國民黨在競選時，對陳水扁政府對中國封鎖政策斲傷台灣的經濟的指控。很明顯地陳水扁執政期間，台灣迅速地向中國開放。

而2007年：

比較2007年與1999年的圖形，我們發現：1)中國經濟迅速地成長（多說的！） 2)台灣對中國迅速開放。台灣與中國的貿易流量高於理論的預測值，我認為可能原因之一是，兩國之間語言與文化相似，降低了許多無法量化的貿易障礙因素。因此，一旦政府在政策上開放，兩國的貿易量便會迅速成長。

最有趣的是這張圖：

我將台灣與中國的貿易量逐年點在圖上，其中的紅線是所有資料點不分年度的配適值。我們發現：1)早期確實有貿易管制的存在：在1992到2000年左右，儘管中國經濟緩慢成長，台灣與中國的貿易流量增加相當地少。 2)2001年之後，中國經濟成長對於台灣與中國貿易量的貢獻增加幅度比其他主要貿易夥伴還大。 3)2004年之後，中國經濟成長對於與台灣貿易量的貢獻似乎有減緩的跡象。可能原因是，中國與台灣的經濟關係已經相當開放，更加開放在邊際上的貢獻有限。或者是，台灣政府對中國經貿開放的腳步趨緩。

不論中國經濟成長對台灣的貿易貢獻減緩的因素為何，我們可以確定台灣與中國之間的經貿的阻礙(friction)已經很少。假設貿易開放確實如綠營所聲稱會對主權會有所傷害的話，而政府又想要以最低的主權傷害拯救經濟的話，已經不應該將中國列在最優先順序。但是與其他國家簽訂類似的經貿開放協議在政治上是否可行，已不在我知識所及範圍。

在 akaruisonne 和我多次討論有關擴大內需時, 最後都趨近演變成新古典和凱因斯的大戰, 而台灣可以看到的主流媒體, 把所有解決當今經濟危機的藥方都開到凱因斯頭上。

我不懂凱因斯, 但是問題的解決方法不是對立的, 很多方法可以解決人類面對的經濟和社會問題, 只是效果好壞. 後遺症多不多. 在某些狀況底下適不適用。更重要的是, 不是『主流媒體』說什麼, 或者是政府現在在做什麼, 就是唯一或有用的藥方。

這是個被遺忘(well, 至少經濟學界的人都還記得他)的經濟學家, Irving Fisher 的故事, 而這個故事, 也是現在 Ben Bernanke 現在在告訴我們的故事。

『The Economist』 proudly represents you:

Out of Keynes’s shadow

最後, 只希望這個社會在任何經濟或公共政策, 可以多聽. 多看. 多採納另外一邊的意見和想法, 這不只是單純的成本效益分析, 更是自由價值的一種展現。

消費券和現金的差別在哪裡？
如果大家拿到了現金都存起來，豈不又掉入「儲蓄的矛盾」、「貧窮的陷阱」之中？沒有儲蓄價值的消費券才能創造現金的流動啊！

正好最近幫妹妹複習大二個體，就用簡單的兩期模型，討論消費券與現金究竟有什麼不同。

為了強調兩者的差異性，我們假設不論發放現金或是消費券，金額皆為3,600元。

一、現金

圖中紅線為補貼前的預算限制。政府發放現金三千六百元後，預算限制變成綠線，因為現金可以儲蓄，所以綠線為紅線向外平行移動。

二、消費券

圖中紅線為補貼前的預算限制，藍線為補貼後的預算限制。由於消費券一定得在當期使用，因此下期的可能消費不會增加，其他部份向外平行移動。

如果將補貼現金和補貼消費券放在同一個圖上比較：

我們發現，發放現金和發放消費券造成預算限制的改變確實有所不同。發放消費券會使得在發放現金的情形下當期消費仍少於3600元的人將增加消費到拗折點（也就是當期消費變成3600元）；而對於獲得3600元現金後本來當期消費就會超過3600元的人，消費券和現金的效果完全一樣！

白話一點說，如果消費券和現金對某甲效果不一樣，一定是因為就算政府發給某甲現金3600元，他在當期消費仍少於3600元，這個情況之下，政府發給某甲消費券能夠強迫他增加當期消費到3600元。如果政府發給某甲3600元現金，他已經願意消費多於3600元，那麼對某甲而言，消費券和現金是完全一樣的政策！

接著，我們可以思考看看，究竟有多少台灣人一個月消費少於3600元呢？更別提，若將消費券使用期限的2009年一月到2009年九月都當作是「當期」來討論現金與消費券的差別了！

最後，既然我們知道現金與消費券政策效果確實有如此微小的差別，「天下沒有白吃的午餐」，考慮發放消費券的成本之後（立委質疑消費券兌換成本高 財部：有檢討必要、發消費券成本超驚人　人事費破3億），不曉得值不值得為了一點的政策效果不同，選擇消費券政策了！

消費券是一種財政政策. 貨幣政策或是福利政策, 在這個時間點的重要性已經不高, 對個人來說, 這個效果與有政府盈餘的退稅不太一致, 而是今天政府發一筆錢給民眾, 而明天(只是這個明天是不久的將來還是明日復明日只有上天知道)國家將會課稅回來, 因此在經濟理論上, 消費券的效果為零, 消費券的金額只要沒有超過一個人的消費總額, 則只是將原本利用現金給付的消費換為使用消費券, 並增加儲蓄等到政府要課稅時將此金額繳還給政府, 在理論上稱之為李嘉圖對等定理 (Ricardian Equivalence)。但在實證上利用日本之前實行類似政策的資料, 消費券的邊際消費傾向 (Marginal Propensity to Consume, 簡稱 MPC) 為 0.1 ~ 0.2, 意指發出每十元的消費券, 增加的消費約為一到兩元。這部分的結果與理論不合, 可能的原因有幾種可能：

1. 由於需要課稅的明日可能是明日復明日, 明日何其多, 如果使用疊代(Overlapping Generation)模型, 如果上一代對下一代效用的『關心程度』(操作上就是折現率)不夠高, 則可能會採取債留子孫的方法來增加自己的消用。

2. 在沒有發放政府補貼前, 個人已經把所有的資源全部投入消費, 則這群人在之前就是處在消費不足的狀況之下, 因此拿到政府的轉移會直接增加消費。

3. 行為上的效果 (Behavioral effects)

這整篇文章的目的, 不是在討論這個政策好不好(當然如果你問我, 我只會笑一笑不知道該怎麼評論), 這個 0.1 ~ 0.2 的 MPC 是否值得, 到最後反而是政治問題, 而不是經濟學問題。但是在這個問題變成一個政治問題之前, 經濟學者有必要先把這個政策可能的成本和效益先分析出來, 外加這個政策的其他可行選項的成本效益分析。

給定政府一定要轉移部分購買力給民間這個前提之下, 那最容易被聯想的對立政策是發現金, 一般對於這個政策的批評在於人民會直接儲蓄, 那以理論上來說, 發給現金與消費券對於一般的消費者並無差距, 但是除了上述提到的第三項效果以外, 其他兩個狀況, 取得現金的消費者也會消費, 因此消費券和現金的差距, 理應就是行為上消費者對消費券和現金的行為差距, 而成本上, 發給現金需負擔的是電子轉帳的成本, 而消費券存在防偽和印製的成本, 再加上家父長『認為』需要正確的『指導』消費券的用法, 所需要的稽查成本, 而這兩者對於消費的增加需要相減, 而並非直接以原點(未做任何政策)為基準, 不然很容易忽略掉其他政策的優缺點。

本篇文章不在於討論消費券政策絕對的是與非, 如同前述, 這些問題在經濟學家的討論已經有定論, 本文只是藉由消費券當例子, 討論公共政策可能被忽略的方向。同時也在看到一堆數字的同時, 知道背後的思考邏輯。