Tag Archives: returns

Deja Vu – Part 3, Lessons from the US economy,1979-1985

26 Dec

In my last blog entry I mentioned three US currency policy decisions over the years 1979 to 1995 (https://ravendragon.wordpress.com/2010/12/16/deja-vu-%E2%80%93-part-2-lessons-from-the-us-economy1973-1993/). Each decision was a response to the decreasing returns in US industry. In this blog entry I will outline each of these decisions in greater detail.

As mentioned the 70’s saw Nixon embark on a policy of federal deficits and extreme monetary ease. One aim of the policy was to lower the value of the dollar in order to sustain the competitive position of US manufacturing. This policy helped the US ride the 1974-5 oil crisis. On the one hand the policy was a great success vis-à-vis US manufacturing. Nonetheless, such extreme Keynesian measures lead to over production and over capacity. In addition, it prevented the much needed exit (what Brenner calls “the medicine of shakeout”) of many lower return companies from the market place. The upshot was an overall continued decrease in returns, which eventually dis-incentivised manufacturers from increasing output. With massive deficits (i.e. increased money supply), and little to no increase in output, the US began to suffer inflation. Simultaneously and related, massive deficits led to run on the USD, threatening its status as a global reserve currency.

Before we see how the US tackled this issue, let’s pause and understand that this dynamic sounds rather similar to the US today. Today the US is running a policy of extreme monetary ease, MASSIVE federal deficits, and has relatively uncompetitive manufacturing compared with countries like China. It led to inflation in the late 70’s. The risks of inflation in the coming years ia fairly apparent wouldn’t you say? In addition I believe today’s policy is aimed at helping US exporters/manufacturers, whether it’s explicitly stated or not.

In 1979/1980 the US enacted the Reagan-Thatcher Monetarist revolution, which reduced corporate taxes, increased unemployment, and removed controls on capital. The aim of the policy was to provide the US economy with a serious dose of “the medicine of shakeout” and purge the US economy of the low profit manufacturing firms that had become an ill of previous Keynesian expansionism, thereby controlling inflation and the currency. However, this time the pendulum had swung too far the other way. Record interest rates and the associated rising USD almost caused a US crash with the threat of contagion.

How did the US (and the global economy) get out of this situation? The government started the Reagan military buildup – purely Keynesian in nature, though most right-wingers would feel queasy to admit it! In addition Reagan enacted a program of huge tax cuts for the wealthy. These 2 policies combined did offset the government’s tight monetarist policies, yet they equally reopened US trade and current account deficits. The deficits required to pay for the military buildup were in fact larger than the deficits in the late 70’s. However, the combination of high US interest rates ensured there was no run on the dollar this time. In fact the currency (aided by a push of capital flows from the Japanese) appreciated considerably.

The Reaganites liked the combination of high interest rates and low inflation. Why? It played to the theme on increased the strength of US financing capital. However, it caused severe pain to US manufacturing. Under extreme pressure from congress, the Whitehouse was forced into a U-turn.

On 22 September 1985 the G5 met to sign the Plaza accords, under which the G5 powers agreed (under US pressure) to relieve the stress on US manufacturing by reducing the USD exchange rate. Over the coming few years the US administration passed a number of pieces of protectionist legislation (including the Omnibus Trade and Competition Act of 1985, and the Structural Impediments Act of 1989). Whilst simultaneously closing home markets to foreign imports the US was forcing its exports and FDI onto overseas markets.

So we see a continued US policy of not dealing with the real structural issue of lowering returns and non-competitiveness. Ford, Nixon, Carter, and even Reagan continued a policy of using the US currency as a band aid to be placed over a brewing problem. We know that since the late 80’s the US has continued in the same vein as we are facing similar structural issues today, and similar short-term currency play and Keynesian remedies are being used by the Feb and President Obama’s administration. As I’ve said before, this pattern of behavior is not a good sign that the US will deal with its structural issues. I will stress again that I am not against a band aid, as long as a real medicine is given in combination.

In the next blog entry I will hope to conclude my economic history lesson. But this is a blog about China, so I will be putting this history in the context of Chinese production and trade. In a nutshell I can already see that the US’s failure to deal with the issue of its lowering returns is a boon for the Chinese, in both the short and long-term future. More next time!

See you all in 2011.

Happy New Year

再见
新年快乐
乔恩 (https://ravendragon.wordpress.com/about/)

 

 

Deja Vu – Part 2, Lessons from the US economy,1973-1993

16 Dec

In my last blog entry I outlined the period of 1950 to 1973 as one of uneven business development between the US and its trading partners Japan and Germany (https://ravendragon.wordpress.com/2010/12/13/deja-vu-part-1-lesson-from-the-us-economy1950-1973/). It was this uneven development, I argued, that precipitated an oversupply of productivity that led to a general decline in returns over 1965-1973. I would like to contend that over the ensuing 20 years (1973-1993), oversupply was in fact exacerbated by both governments and businesses alike.

Theoretically, the market lives under an evolutionary arms race – survival of the fittest. In other words, when profits shrink, it should present a barrier to new market entrants, and ‘kill’ those companies with a cost of capital higher than their returns. However, in reality such spontaneous mechanisms do not always exist. Higher cost incumbents resist leaving the market, as their assets (both tangible and intangible) cannot switch lines. Instead, high cost producers tend to invest further capital in improving their productivity. This in turn creates an impetus for higher return companies to follow suit, which creates a cycle of lower profits/returns. Put simply, there’s no spontaneous mechanism for companies to exit the market!

On the other side, a fall in profitability precipitates the search for lower production costs. Countries and regions with lower labor costs are thereby incentivized to join the production band wagon. Over the past 2 decades Asian economies (especially China) made rapid inroads into the manufactured goods market. Yet such new entrants simply push down the profitability and returns of the industry even further. Put simply, as returns fall, there’s no spontaneous mechanism to prevent new market participants!

Over the years 1973 to 1993 world governments continued to deal with the issue of reduced returns with currency revaluations, much as we saw in my last blog entry. In 1979-80 the world witnessed the Reagan-Thatcher Monetarist response, which reversed the devaluation of the USD made during the 70’s. The 1985 Plaza accord resumed a policy of devaluing the dollar. In 1995, the US government once again reversed its policy of devaluation.

But before we get into the details, remember that the purpose of this series of blog entries is not a history lesson or political polemic, but to try and see parallels with today’s global economic, and assess the likely response of our governments over the coming years.

Without yet delving into details, I can already see that despite a serious structural issue leading to reduced returns, world governments did not contend with the core issues. Instead, they perused short-term policies aimed at plugging the hole, mainly in the guise of currency games.

This does not bode well as a precedent for dealing with today’s issues: the US “balance sheet” and the trade imbalance between China and the US.

Let’s not get ahead of ourselves though. In my next entry I will discuss what led to those three currency policies over 1973 to 1993, and the impact of each of those policies. Armed with that information we will be much better armed to try and shed some light on the coming few years.

 

Note – It’s been a while since I wrote in my series on building a core China portfolio (https://ravendragon.wordpress.com/category/building-a-core-china-portfolio/), but I hope to return to it sometime in the new-year.

Zai Jian

Deja Vu – Part 1, Lessons from the US economy,1950-1973

13 Dec

I remember a conversation with a colleague in the office kitchen one Monday afternoon. The conversation occurred during my first banking job over a decade ago. I asked “How was your weekend?” to which my colleague replied, “My boyfriend took me to the Imperial War Museum over the weekend. They had something on about Musso, Musso, Musso… some Italian bloke.” “Mussolini?” I interjected, “Yes…” she said “…that’s the one. I was never one for history. Don’t understand the point in it. It’s all dead and gone now”. We had a brief discussion about the importance of history giving identity, and the opportunity to learn from trends and mistakes. At the time it struck me that this lady had just as many votes as me in a general election (only kidding!) Nonetheless, over 10 years on I decided to take my own advice and use history to understand some of the dynamics presently at work in global economics, and to use this information to help understand what is likely to happen in the coming years.

Analysts, the press, and even my own blog have been dealing with the issues of the trade imbalance between China and the US, the value of the Yuan vs. the USD (and the potential for a currency war), and the aftermath of the recent credit bubble/crisis. Yet, none of these issues are new. As recently as the past 50 years global trade and economics has played out the same script. The same questions have been asked in the lifetimes of people I sit amongst, albeit between the US on one side, and Germany and Japan on the other. I’ve not seen these precedents referred to by many experts, or by industry colleagues, so the first lesson I’ve learned is that we are very quick to forget, which means are most likely damned to repeat. But let’s not get philosophical yet! Let’s first outline the history.

The “West” enjoyed a long period of boom over the 50’s and most of the 60’s. The boom itself, however, was rooted in a fairly uneven development between the US, Japan, and Germany. Through a combination of “importing” American high-productivity technological advances and domestic low cost elastic labor supplies, Germany and Japan were not only able to catch up with the US in terms of growth, investment and returns but often surpass the superpower.

Until the mid 60’s this was all in the US’s favor. Firstly, it provided new markets for US capital. Secondly, it provided new markets for US goods as Japanese and German consumers prospered (note – at the time it was relatively harder for Japanese and German goods to enter the US market). And thirdly, it was a key component in fighting the cold war.

By 1965 things began to change. The economic development of Japan and Germany presented real competition to the US. Their growth was beginning to become a zero-sum-game (or even negative sum game) from the perspective of the US. During the years 1965-73 Japanese and German exports began to interrupt the easy flow of US goods onto global markets. In addition, their lower cost goods (of equal quality) were increasingly making their way into the US market. The impact was to decrease the return on capital in the US.

The US responded by lowering prices (i.e. lowering their returns to the cost of capital), capping wage growth, and updating plants and machinery in an attempt to produce more efficiently. However, its greatest weapon was the devaluation of the dollar.

Between 1969 and 1973 the dollar devalued by around 50% against the German Mark, and between 1971 and 1973 by about 30% against the Japanese Yen. How?

In the early 70’s the US followed a combined policy of lower interest rates, a policy of easy credit, and eventually the total abandonment of fixed currency rates in favor of a free floating currency by 1973. As the US government pursued a policy of fiscal stimulus in 1970, interest rates fell, and the Fed abetted with a policy of easy credit, I am struck by Nixon’s comment at the time that “We are all Keynesians now!”

Granted the story is a little more complex in terms of which parts of policy were a priori thought out, and which parts were responses. The balance of payments, impacts on the stocks markets, and inflation in Japan and Germany cannot be ignored. Nonetheless, the general picture of 8 years of economic struggle (to borrow a phrase from David S Landes) over 1965-1973, the reduction in the competitive position of US returns, and the Keynesian and currency devaluation responses are clear.

The devaluation of the dollar led to a turnaround in US manufacturing sector that restrained wage growth and productivity growth had blatantly been unable to achieve.  Profits, investments and productivity returned to US manufacturing, and the trade imbalances with Japan and Germany closed. The impact in the Japan and Germany was the opposite, as they sought to keep sales by suffering lower returns.

I will return with a second installment of this story in the coming weeks, and a closer of analysis of the meaning of these events for today’s economy and global trade. Nonetheless I want to end with a couple of points to consider already:

  • We don’t need all of the subsequent history to begin to see parallels with the state of the US economy in 2010 and the early 70’s, as well as parallels between modern US and China relations, and historical relations between the US and its trading partners in the 60’s.
  • I want to leave you with an additional question. Why did the capitalist system result in such a decline in US productivity and returns over the period we’ve discussed? Is the capitalist dynamic insufficient when it comes to ensuring the rejection of waste and barring the entrant of excess? We will return to this issue as well.

Building a Core China Portfolio – Part 4, Let’s Start Digging!

5 Sep

In my previous blog entry (https://ravendragon.wordpress.com/2010/09/02/building-a-core-china-portfolio-%E2%80%93-part-3-assessing-quality/) we discussed the notion that a lot can be understood about a company by its numbers, before we even know the details of its business model and drivers, let alone speak to the company’s representatives. We are using VTech Holdings as our first example based on the 100 companies delineated in my shortlist to be found at the following URL: https://ravendragon.wordpress.com/2010/08/25/building-a-core-china-portfolio-%E2%80%93-part-2-the-shortlist/

To reach a deeper understanding we will compare the ROE of VTech with its main competitors, as well as the components of said ROE with that of main competitors.  But that leads us to the following question: who are VTech’s main competitors?

The company reports 3 main segments:

  • Manufacture and supply of corded phones (TEL)
  • Electronic Learning Products (ELP)
  • Contract Manufacturing Services (CMS)

Over the past 5 years the TEL segment has contributed to about 45% of the company’s top-line, and the ELP segment to just shy of 40%. As such, I consider 2 peer groups to be useful:

  • Phone manufacturers
  • (Electronic) toy manufacturers
  • [One could potentially look at VTech in the context of Hong Kong exporters as about 55% of revenues come from the US, and 35% from the EU]

Before we start digging into the toy manufacturers, let me just remind you of the components of ROE so you can understand why I am performing the decomposition below, ans see the influence each component has over the final return figure:

ROE = [EBIT%]  x [Asset Turnover] x [1-Tax Rate] x [Leverage]


Toy Manufacturer Data

Figure 1 – Historical ROE vs. Toy Peers

Figure 2 – Historical EBIT Margin vs. Toy Peers


Figure 3 – Historical Asset Turnover vs. Toy Peers

Figure 4 – Historical Leverage vs. Toy Peers

Now let’s move onto the phone manufacturers:

Figure 5 – Historical ROE vs. Phone Peers

Figure 6 – Historical EBIT Margin vs. Phone Peers

Figure 7 – Historical Asset Turnover vs. Phone Peers

Figure 8 – Historical Leverage vs. Phone Peers

Interpretations

For most of the past 2 decades VTech has been the best in class performer in terms of ROE in both peer groups. This is evident from figures 1 and 5. That’s no simple accomplishment. Let me assure you, that this is a live investigation – I’d not seen this data up until this point. Let me also remind you that consistency and sustainability of strong results are they key element we are looking for.

Figures 3 and 7 show that the asset turnover (sales/total assets) at VTech is simply world class, and has contributed heavily over 20 years to the company’s impressive returns. In a coming blog, if we decide to stick with VTech we will want to break this down and see what is contributing to the strong asset turnover (e.g. is it heavily influenced by fixed asset turnover, net working capital turnover, and are there any weak links).

Normally one finds that companies with high asset turnover have lower EBIT margins, and vice-versa.  It’s normally a reflection of whether the company’s strategy is more focused on volumes (i.e. high asset turnover) or high pricing (i.e. high EBIT%). With that in mind we can perhaps understand why VTech’s margins have been historically lower than its toy peer group (see figures 2 and 6).

In fact looking at these figures we see that it was a negative EBIT margin in 2001 that significantly contributed to the poor ROE in that year (it was exacerbated by a slightly lower leverage in 2000-2002).  If we continue beyond today’s blog entry with VTech, we will want to understand what happened that year and ask questions such as the following: Is the company more prone to cyclical risks than its competitors in both peer groups? If so why? Has this changed? How? Is it simply more operationally leveraged? Again why? Etc… Was this a one off?

Despite this weaker long-term margin history, I am very encouraged to see that EBIT margins performed very strongly in both absolute terms and relatively compared with both peer groups (see figures 2 and 6) since 2006. Is this sustainable, and if so how? What changed? Was it simply mix? Geography? Etc… We will want to find this out.

Conclusions

This is a quality company. It is worth understanding the business model, drivers, and future sustainability.

  • Strong sustainable ROE (both relatively and on an absolute basis) of about 40%
  • driven mainly by very high asset turnover
  • and good EBIT margins in recent years.
  • Leverage is within the peer group ranges and at very acceptable level historically.

Next I will be asking questions to dig into qualitative elements of the firm in order to judge its sustainable profit levels, provide an upside / downside valuation range.

With such a value range we will be able to understand a good entry point for VTech and judge whether the current risk reward posits a buying opportunity.

Caveat

One point at this stage – I notice that the company does not report product segments below the revenue line. I do not like this as I am unable to ascertain the strengths, returns, and sustainability of each part of the business, and to know whether one is dragging the other up or down. I will look more at this.

Zai Jian

Building a Core China Portfolio – Part 3, Assessing Quality

2 Sep

Our aim is to find companies that fit the following bill:

  • high quality
  • low balance sheet risk
  • no-left field risk
  • quality of earnings (i.e. cash conversion)
  • favourable capital allocation vis-a-vis equity holders
  • strong competitive position
  • solid drivers

If a company does not fit our profile, we reject it and move onto the next one. If it does meet our profile, we can move onto estimating sustainable income and putting an upside/downside value range on the firm. With comfort about the quality and risks, and a fair value range (with a high margin of safety) one can decide whether now is the right time to purchase the stock, or whether to keep monitoring (and amending with new information), and waiting for the right moment to pounce.

Now, time is our most precious commodity.  With that in mind I start my list above with an assessment of quality and balance sheet risk. It is fairly easy to get a quick assessment of these 2 criteria, and know whether it’s worth spending further time on a deep understanding of the company’s business model and working on a valuation.

The measure of quality is returns. As a potential equity investor one must ensure that the company has a good track record of returns on its equity (ROE).

Yet how does one define quality. The first basic benchmark is to look for companies with an ROE of at least 15%. This is not fool proof, and will often be industry dependent. However, on average the market is returning around 12%, and with many companies beating that average, below 15% represents a healthy level at which to reject a company with comfort in the knowledge that there’s still a huge universe to choose from.

(Note – It usually yields greater rewards to find quality and wait for a good entry price rather than go for low quality at a very discounted price. It’s just common sense that sustained low quality is less likely to improve than a discount opportunity is likely to close on a company with sustained quality).

The second thing to ensure is that a company exhibits a sustained (or growing above the benchmark) return over a period of many years. A track record is important for obvious reasons. I will add though, the success of a company is clearly heavily dependent on management. With one of managements’ primary task being the successful allocation of capital in a firm (in order to create value i.e. grow its returns above the cost of said capital),  a good track record of sustainable returns is a great yardstick for the competence of a management team, and with all other things being equal, is a good indicator of future success of the teams execution (this naturally has its caveats and should not be relied upon in isolation).

The next thing to do is chose a number of peer companies and compare how our company under analysis has performed compared with the competition. If the company has severely underperformed I suggest rejecting it at this early stage and moving on.

If the company has:

  • had only a couple of poor years
  • had a sudden jump or drop
  • performs relatively well compared with peers
  • or has even performed well or very well

…then it is worth breaking down the components of its ROE to understand what’s driving that performance and or blips, what’s causing the differences etc… and then making an initial decision about the company’s quality.

What are the components of the ROE?:

ROE = [EBIT%]  x [Asset Turnover] x [1-Tax Rate] x [Leverage]

(Note – this is ROIC x leverage)

Anyway, this blog entry has been very theoretical. In the next entry we will apply these ideas, present and have a close look at some of the data.

I want to leave you though with something more concrete.  In my last entry on building a core China portfolio (https://ravendragon.wordpress.com/2010/08/25/building-a-core-china-portfolio-%E2%80%93-part-2-the-shortlist/) I presented a short list of 100 companies that deserve further inspection (https://ravendragon.files.wordpress.com/2010/08/china22.gif).

We decided to start looking at VTech Holdings (HK 303). Below is a graph of VTech’s ROE between 1993 and 2010.

On first inspection it looks great. It increased steadily from 1993 to 1998 to high 30%’s (easily above our benchmark 15%). Then it collapsed when the tech bubble burst (will need further investigation). Subsequently, the company has posted returns sustainably in the high 30%’s, reaching 50% for a couple of years. Even 2009/10 was very strong.

Conclusions thus far – let’s keep looking at Vtech!

Building a Core China Portfolio – Part 2, The Shortlist

25 Aug

The table below contains the top one-hundred companies shortlisted according to the criteria outlined in part 1 of my “Building a Core China Portfolio” series, ranked by best return over a ten year period.  (Find part 1 at – https://ravendragon.wordpress.com/2010/08/22/building-a-core-china-portfolio-part-1/)

Before I explain some of the details, I know that the table is too small to read. All you need to do is click on it and it will open as a stand alone table in your browser. Next simply enlarge the table (clicking on the table itself is one way to zoom in) to a readable size. Alternatively you can right click on the table to open it in a new tab or window and similarly enlarge.  Either way, I am happy to be contacted and email you the table in excel format.  Send requests to raven.jonathan@gmail.com

It will take me a little time to clean up, but I am planning to set up the table for users to be able to open a dynamic excel spreadsheet with their bloomberg terminal. If you would like a copy of the dynamic table when I’ve fixed it up, please contact me at the same email address.

Anyway, back to the table itself.  The table contains the top 100 companies ranked by 10-year returns. In addition, I excluded all companies with a maket cap below 5bn in local currency in order to reduce and element of trading-liquidity risk a this stage.

The table contains the following data points:

1. Company name

2. Bloomberg Company ticker

3. Market Cap in local currency (HK$s or CNS)

4. p/e for current and next year (so we can see whether we are getting a good price for the quality we’ve identified in the company’s returns – a lot more on this in coming blog entries).

5. Capital employed (i.e. increase in common equity and net debt), increase in net income, and return over 5, 10, and 15 years.

So what comes next?

Remember, I’ve said that this is a live investigation, and I’ve no idea where is will lead or end. Well next I simply plan to work my way through the list from the top to bottom, choosing which companies are worth expending my time on a thorough investigation.  Time is afterall our most precious commodity.

Let’s look at the top three companies to begin with. Clearly Vtech Holdings is the place to start an investigation.  Not only does it have a 15 year record, but it also had very strong returns over all three periods. In addition, the p/e of 9.4x 2011e bloomberg consensus earnings is a good price for a company with such strong returns.

So, in part 3 of this series I will take a closer look at Vtech. The discussion on Vtech will also present an opportunity to talk about some methodology.

Zai Jian – Jon

再见 -乔恩

Building a Core China Portfolio – Part 1

22 Aug

I’ve wanted to build a China portfolio for a while.  But I’m not (yet!) an expert on Chinese stocks.  So in order to make an educated decision (and improve mine and your expertise along the way), I’ve decided to apply the same principles I’ve learned and used in other markets.  My goal is to discern which companies should form the core of my portfolio. I’m not sure where my investigation will take me, but I invite you along for the ride.

The ultimate goal is to look for quality investments (sustainable high returns) with no balance sheet or left-field risk, and a competitive edge.  This must all come at a reasonable price with a wide margin of safety.  These are the kind of names one holds for the long term i.e. investments, not speculation.  And yes… such companies do exist!

The first step is often the most difficult.  With 1000’s of mainland and Hong Kong listed companies, where does one start?  Well I decided to apply a basic return heuristic to the entirety of the Shanghai and HK stock exchanges, and rank for the best performers – i.e. create a shortlist for deeper analysis and investigation. I admit that it’s not a fool proof system, and I know quality companies will fall by the wayside at this early stage, and that low quality might slip through the safety net. But the logic is fairly sound, which should lead to a fairly sound short list of 10’s of stocks.

I’m going to take every listed company and compute a form of return over a series of different length periods. The reason to take different periods (5, 10 and 15 years in this study) is to test how robust the returns of the company are historically. Sustainable returns are important as returns are the most important indicator of quality to a stock analyst.  If we understand the environment (both internal and external) that contributed to said returns, and can qualitatively assess how likely those factors are to continue in the coming years, that combination of solid facts and common sense is as close to science as stock picking comes.

In addition, the primary job of management from our perspective is to allocate capital effectively within their company and create value for stockholders. In other words, their job is to consistently deliver returns greater than their cost of capital.  Many analysts will meet management and declare whether or not they are good or bad.  Now my first degree is in psychology, but it didn’t make me a mind reader. How can an analyst jump to such conclusions, often using them as a basis for his or her investment decision? I don’t know how. But the return data does not lie, and solid returns over an extended period (note – solid depends on many things, and will vary according to sector – though as a rule of thumb, the developed market is returning about 12% return on book equity) is a good indicator that management knows what its doing. This data point should not be used in isolation to judge management, but the importance of capital allocation and returns to the shareholder, along with the dependence of returns on management decisions, makes good management golden.

Anyway, back to creating our shortlist.  The return heuristic I am going to calculate is the following:

Increase in net income over the period / (Increase in common equity + increase in net debt)

In other words, this tells me over a certain period of time how much capital a company employed to create additional income over that same period. The higher the return sustainably, the better the company has been at using its additional capital to create value for its shareholders. Granted it uses P&L data which is itself riddled with caveats, (not to mention that the system does not account at this stage for capital structure) but ranking such returns over different periods and assessing sustainability is a good way to create a short list of 10’s of companies for close inspection from a universe of 1000’s.

I will use Bloomberg data, again not fool proof, and I caution against making a final decision to buy (or short) without reading or modeling from numbers taken from actual annual reports. But I’m happy using their data as part of an exclusion exercise, as it is mostly robust and extremely reliable

I will publish my findings in the coming days.  Maybe I will put some EU and US data so that readers familiar with those markets might feel more comfortable.

Zai Jian