ࡱ>  i !hbjbj .nUf%:M,E,E,E,Y,Y,Y,8,-Y,a01"111دگگگگگگ$YE,>^yy11sh̬̬̬~y.1E,1د̬د̬̬%̬1/Y,p̬į10a̬̬E,̬̬Ja : Crises and recovery in emerging markets: Phoenix miracles or endogenous growth? Marcus Miller a,b and Lei Zhanga a University of TV and CSGR b Centre for Economic Policy Research, London January 2007 Abstract The Great Depression of the 1930s gave birth to Keynesian economics with its focus on aggregate demand. In recent crises in emerging markets, currency devaluation and associated balance sheet effects have played a key role. Does this call for a new paradigm? After a comprehensive survey of many crisis episodes - including the US Great Depression - Calvo and colleagues from IDB offer the paradigm of a Phoenix Miracle. It is a static, supply-side account where factor productivity falls in recession and rises promptly thereafter. By contrast, we argue that the real miracle in East Asia is economic growth. This is rudely interrupted by external shocks: but then it restarts at a lower level of trend GDP. This we show in a simple model of endogenous growth, disturbed by shocks to supply and demand. Heterogeneity of production implies that both are relevant - as in New-Keynesian paradigm for monetary policy proposed by Stiglitz and Greenwald, where monetary contraction affects both supply ( of traded goods) and demand (for non-traded goods). Introduction Financial crises in SE Asia sharply interrupted stellar economic progress in the region: for a year economic growth went into reverse. It did not take long for growth to resume, but output has followed a lower trend path. What happened, and why? That these events were triggered by external capital market shocks is now widely accepted; in the terminology developed at the IADB, events in SE Asia followed upon a Systemic Sudden Stop in capital flows. Liability dollarisation and the balance sheet effects that accompanied currency devaluation undoubtedly played a key role in depressing investment and supply, as did the high interest rates used to defend the currencies. There is, so to say, agreement on the importance of the Fisher effect, a central element of so-called third generation models of crisis. But how to explain the sharp falls of output that were almost synchronous with the devaluation - and prompt recovery thereafter? Here there are sharp differences of view. Some argue that the conventional stimulative effect of devaluation in the Mundell-Fleming model of a small open economy must have been offset by the (temporary) adverse demand effects of liability-dollarisation characteristic of emerging markets. After a comprehensive overview of crises in SE Asia and elsewhere, particularly in Latin America, Guillermo Calvo and colleagues at the IADB have arrived at a different conclusion. The fall of output was attributable to adverse supply-side effects sharply lowering total factor productivity, and its prompt reversal is what they call a Phoenix miracle, CIT(2006). It is not a matter of remembering Keynes: it is understanding how supply contracts in a financial crisis. To capture the setting of ongoing growth in SE Asia, we first examine these issues in the framework of a simple model of endogenous growth. In this context, the Fisher effect certainly interrupts growth but does not in itself cause a sharp fall in output. A temporary fall in TFP will cut output, of course: but so too does a temporary fall of output below existing capacity. The endogenous growth model used is of a closed economy, but the arguments go through in a model of a small open economy with a Fisher effect, as shown in the following section using the popular 3G model of ABB. Shocks to demand and supply may be may be observationally equivalent at an aggregate level, but not when account is taken of the distinction between traded and non-traded goods. Supply side effects will be more relevant for traded goods and demand effects for non-traded, as in New-Keynesian paradigm for monetary policy proposed by Stiglitz and Greenwald, where devaluation and tight money can restrict aggregate demand and disrupt supply in the traded goods sector (where demand is not a problem). 1. Literature review The deterioration of corporate balance sheets is a key element, as is clear from so-called third generation models of crisis. But the role of debt in triggering economic contraction has a much longer pedigree: what Irving Fisher referred to as the debt deflation mechanism operating in the US Great Depression provides a telling precedent. But there is a key difference. For Fisher balance sheets deteriorated as falling prices raised the real value of nominal debt: while in emerging markets it is devaluation against the dollar that raises the local currency value of debt contracts denominated in foreign currency. In the Biblical terminology of Eichengreen and Hausmann, the crushing impact of the Fisher effect on private enterprise in emerging markets is a legacy of Original Sin. In Latin America, for example, the fear of floating described by Calvo and Reinhart (2002) may be attributed to such balance sheet effects of devaluation and to the high interest rates implemented to check the fall in the currency. Keynesian models can be adapted to take account of the Fisher effect. In is analytical afterthoughts on the Asian crises, for example, Krugman (1999) used a demand-side account of a small open economy to argue that a loss of confidence by foreign investors can be self-justifying, because capital flight leads to a plunge in the currency, and the balance-sheet effects of this plunge lead to a collapse in domestic investment. In a much more detailed framework, Cspedes et al. (2003) show that for a highly dollarised open economy, the asset price effects of devaluation can overwhelm trade effects, leading to a contraction of aggregate demand. In the third generation approach developed by Aghion, Bacchetta and Banerjee (2000), however, it is supply-side effects that play a central role. In the first place, the trigger for crisis is an unexpected idiosyncratic, permanent fall in productivity which lowers expected future supply. In the absence of a corresponding contraction of future money, the currency is expected to be weaker in future. Anticipation leads to prompt and unexpected devaluation with adverse balance sheet effects on investment further reduce future supply. Current output is unaffected by the crisis, however. Productivity effects also play a key role in account of Calvo et al (2006), hereafter CIT, though here they are endogenous. The causal factor is a Sudden Stop which leads to devaluation and a fall in productivity of currently installed plant and equipment. Sharp nominal (and real) currency devaluation in the presence of Liability Dollarisation may have worked in Emerging Markets as a new version of Fishers Debt Deflation syndrome, and may be central in explaining output collapses. CIT (2006p.10,11). How this might occur is explained by a partial equilibrium model, where all output can be sold, but a sharp rise in the ex ante short-term cost of borrowing not only reduces inventories and but also induces the firm to sell physical capital to finance inventories: so firm output falls. Happily these effects are temporary and productivity soon recovers without recourse to outside finance for the firm: this is the Phoenix miracle. [Picture of the U-shaped curve.] In Towards a New Paradigm in Monetary Economics, p.264, Stiglitz and Greenwald stress both the demand-side and supply-side effects of tight monetary policy at the time of the crisis. In the case of Korea, for example, where interest rates were pushed above 25% to try to stabilise the currency, they argue that: it was exporters failure to respond to the huge exchange rate reduction which should have stimulated demand - which makes it clear that the economy was not just responding to a fall in aggregate demand. Such consequences were inevitable, unless the producer could obtain cheaper credit elsewhere or unless wage and price adjustments were sufficiently large to compensate for the huge increase in capital costs. In practice, even reductions in real wages of 20% or more did not suffice. Thus monetary policy had the usual effect on aggregate demand (amplified by the adverse effect of f increased bankruptcy probabilities on firm demand) but they also had huge effects on aggregate supply. [More here] Views on the role of financial effects and the transmission mechanism are summarised in the following Venn diagram. The papers just discussed appear in the left circle as they ascribe a key causal role to the financial shocks (FS) sharp increases in net liabilities amplified by high interest rates used to defend the currency; but they differ on the transmission mechanism whether they work through demand or supply. Figure 1 Financial shock, demand and supply  SHAPE \* MERGEFORMAT  An interesting contrast is provided by the treatment of the small open economy in Obstfeld and Rogoff (1996, Chapter 10.2), where the output of non-traded goods, whose prices are fixed in the short run, varies with the exchange rate. Monetary expansion leading to unanticipated devaluation has an unequivocally positive impact on demand and production of these goods as they become cheaper than flex-price traded goods. Maybe because it was written before the Asian crises, there is no role for the balance sheet effects in this model. Much more emphasis here: To analyse these issues without taking into account the ongoing economic growth in the region; the miracle that CIT report is based on a static analysis This is indicated briefly in the next section where the U shaped curve of CIT is shown alongside with the long run growth. East Asian countries may have suffered sharp, V-shaped recessions; but vigorous growth resumed therafter. In the analysis that follows, a simple AK model of endogenous growth is used to capture the ongoing process of rapid expansion and its interruption due to the Fisher effect, be it mediated through reduction of demand or supply. II. Phoenix miracle; or interrupted growth? In CIT (2006), the stylized facts characterizing 3S output collapses are presented in a series of fascinating graphs averaging data across affected countries. To illustrate, Figure 1 reproduces the typical path of GDP so derived; and a matching index of the capital stock.  These episodes, they observe, are characterized by two salient features. First, there is a dramatic collapse in output (the average fall in GDP is 10 percent) accompanied by a collapse in credit, but without any correspondingly sharp collapse in either physical capital or the labor force. Second, recovery to pre-crisis output is swift and credit-less Thus, although a credit crunch appears to be central for explaining output collapse, recovery can take place without credit. This remarkable phenomenon that resembles the feat of the proverbial bird rising from its ashes prompted us to call it Phoenix Miracle. A longer run perspective may be obtained by plotting GDP for some time before and after the crisis. The longer run of data allows one to fit two trend paths, one before the crisis and one after (with quarterly seasonal adjustment around the split trend), so as to provide a crude estimate of potential GDP, of long run supply.  For the case of Korea shown in Figure 2, starting in the box showing data around the time of the SE Asia crisis, one sees that GDP follows broadly the same trajectory shown in the earlier figure. Looking outside the box, however, gives a new perspective: though the trends fitted before and after have much the same slope, there is a difference of about 10 percent in the level. From a growth perspective, therefore, there appears to be a permanent loss of potential output. (Figures for Thailand and Malaysia given in the Appendix show a similar pattern of a down-shift of trend potential.) In addition to the sharp fall in output summarised in Figure 1, there also appears to be a down-shift in the trend path of output growth. As to why this might be so, the behaviour of capital stock shown in Figure 1 provides a clue: on average net capital formation effectively ceases during the crisis. In the next section we sketch an endogenous growth model where a temporary dip in GDP leaves a permanent mark on potential GDP. The data for output in India shown in Figure 2 provide a striking contrast, with very little evidence of the shock that shook its Eastern neighbours. According to Williamson(), the main difference lay in the degree of capital market control exercised there at the time: so the Indian experience offers support to the view of CIT that the origins of the crisis lay in a Sudden Stop to capital markets. III. Supply, Demand and Crises: an AK approach As Fischer (2001) emphasises, financial crises in South-East Asia involved an initial reversal of the growth rate, followed by prompt recovery: so output traces a V-shape and as shown in Figure 2. These output effects can be explained in stylised fashion using a simple AK model of endogenous growth incorporating balance sheet effects and their impact on productivity and/or aggregate demand. Let potential supply (Q) be determined by the capital stock (K) so  EMBED Equation.DSMT4  (1) and net capital formation be defined as  EMBED Equation.DSMT4  (2) where  EMBED Equation.DSMT4 is the forward difference operator, and  EMBED Equation.DSMT4 is the rate of exponential depreciation. Gross investment depends on the flow of savings and the impact of balance sheet effects, so  EMBED Equation.DSMT4  (3) where s is the propensity to save and  EMBED Equation.DSMT4  is an adverse balance sheet effect, a Fisher effect. In the absence of balance sheet effects, one obtains the canonical growth rate, namely:  EMBED Equation.DSMT4  (4) Phoenix Miracle If an adverse balance sheet effect were simply to cut investment by K for one period, the growth of capital and output following the shock will simply fall by , so  EMBED Equation.DSMT4 ; (5) but there will be no fall in output at the time of the shock. Adding a temporary fall in productivity at the time of the shock, so EMBED Equation.DSMT4 , will reduce savings so capital will grow more slowly after the shock, specifically  EMBED Equation.DSMT4  (6) Further, the fall of productivity means that output growth going into the recession will be approximately  EMBED Equation.DSMT4  And there will be a sharp recovery coming out, namely  EMBED Equation.DSMT4 . Consider for example the case where the balance sheet effect just offsets canonical growth, i.e.  EMBED Equation.DSMT4  , but the productivity effect is twice as large, i.e.  EMBED Equation.DSMT4 . In this case output growth will fall to gc on entering into recession but recovery will take place at almost twice the canonical rate. This is illustrate in Figure 6.  SHAPE \* MERGEFORMAT  Figure 6: Financial shock: short and long run effects on supply. The impact of an adverse balance-sheet effect induced by currency devaluation at period 1, assuming no productivity shock, means that output is not affected in period 1 but falls below the pre-existing trend by  EMBED Equation.DSMT4 % in all subsequent periods, as shown by the upper solid line in the figure. If the balance sheet effect triggers a fall in TFP so that the growth rate of GDP changes sign (as was roughly the case for countries in SE Asia), then the dip will lower potential supply yet further by the amount, (gc+) even if productivity recovers promptly in period 2. With the productivity recession, the lower bold line shows output exhibiting the familiar V-shape in the period of devaluation, and recovering promptly thereafter (but to a lower trend) in period 2. This satisfies the ouput pattern of a Phoenix miracle and the longer term downshift of supply shown in Figure 2. As Talvi (2006) indicates, a similar pattern may be observed in recent crises affecting countries in Latin-America. Using Central-American GDP as proxy for the trend of potential supply, he finds the characteristic V-shaped recession. Economic recovery, accompanied by the redistribution of income in favour of profits but without external credit, is fairly rapid; but it does not take output back to the previous trend. Note, however, that much the same results follow if recession causes demand to fall below supply. Keynesian Recession  SHAPE \* MERGEFORMAT  Figure 7: Financial shock: effects on demand and supply: opening up a gap of If there is demand falls so that income falls 5% as the economy moves into recession as a result of an adverse balance sheet reducing investment by bK,. As before an adverse balance sheet effect which cut investment by K for one period but left demand equal to capacity, would cut the growth of capacity by  for one period as before, with no fall in output at the time of the shock (as consumption must take up the slack). But what if demand Y falls beneath capacity Q as a result of the shock to investment so an output gap opens such that  EMBED Equation.DSMT4 . Assume that investment is otherwise unaffected but consumption falls by sbk. For simplicity that consumption depends both on capacity output and on current income, specifically C= (1-s)Q - s(Q-Y) = Q -sY So Y = C +I = Q-sY + sQ-bK =(1-s)Q sY -bk So (1-s)Y = (1-s)Q-bk So Y=Q-bk/(1-s) If we assume that bK is such that the effect on ouput is the same as it was in the miracle case then the effect on output in the next period will be the same as well. (Note that aggregate demand will fall by (1+s)bK =Q, which implies that bK =is Q/(1+s) i.e. somewhat smaller than above, the reason being that the gap and the investment fall have to satisfy a multiplier relation.) be approximately  EMBED Equation.DSMT4  And there will be a sharp recovery coming out, namely  EMBED Equation.DSMT4 . Do another with full multiplier B Financial shocks in a small open economy We leave longer run growth aspects to one side, to consider how financial shocks might have both supply side and demand effects. In an open economy financial restriction could, for example, could affect the supply of traded goods (for which there is elastic demand), and the output of non-traded goods (where it is only local demand that matters). These are the lines along which we interpret ABB(2000), a popular model of monetary policy in a small open economy written to capture to the exchange rate and balance sheet effects operating in East Asia. Their analytical framework is used, subject to two changes inspired by the work of Calvo and his colleagues at IDB. First the cause of the crisis is taken to be an external financial shock (not an exogenous, domestic productivity shock). Second financial conditions are assumed to impede the expansion of exports in the short run so that the economy lacks the stabilising feature of unlimited foreign demand that assure full employment of resources even in the short run. The results can be interpreted as the effect of a Sudden Stop in capital flows; in an economy temporally deprived of an automatic stabiliser. Follows a short section which starts with credit crunch and has tight money causing a deep crisis because exports dont respond for supply side reasons. Compare this to Greenwald and Stiglitz. Conclusion Not difficult to explain the downshift in trend path - not as a productivity shock but as a result of lost investment.  EMBED Equation.DSMT4  (6) There are different views on how to explain the prompt fall in output. On the one side there is the Standard Keynesian account which stresses the role of demand what Temin(1976) calls the spending hypothesis (and to which he attributes the sharp downturn in the US from 1929 to 1932). It is more plausible that production of nontraded goods could be demand constrained than for traded goods. But, note that, if the price of traded goods moves with the cost of foreign currency, the relative price effect will tend to increase demand of non-traded goods, Obstfeld and Rogoff(1996). So it will require falling incomes of consumers to reduce consumer demand, this could come from a squeeze on real wages (a charactersitic of recent crises noted by CIT), or as knock-on effect of the fall of investment demand, which will be hit directly by the Fisher effect, as ABB point out. By contrast, CIT discount the demand-side completely in favour of an inventory-driven account of a Phoenix miracle. Firms can sell anything they produce, but high interest rates persuade them to economise on inventories and to sell capital. The partial equilibrium account they offer is somewhat problematic as it requires that physical capital is perfectly liquid in a recession and fails to explain why reduced output in the firm selling capital is not largely offset by increased output from the buyer. It takes no account of liability dollarisation, moreover, nor of how a substantial shift in the distribution of income from wages to profits might help produce a creditless recovery. By contrast the New Keynesian approach taken by Stiglitz and Greenwald allows for supply side effects on suppliers of traded goods, as well as aggregate demand effects on the economy as a whole (mainly non traded, presumably). In addition the emphasis on bankruptcy risk allows a direct role for the Fisher effect. The benefits of this eclectic approach is that requires no miracle for output to recover: rising demand, falling interest rates, banks getting back to business and receding threats of bankruptcy will do. References Aghion, Philippe, Philippe Bacchetta and Abhijit Banerjee (2000), A simple model of monetary policy and currency crises, European Economic Review 44(4-6), 728738. Aghion, Philippe and Abhijit Banerjee (2005) Volatility and Growth.Clarendon Lectures in Economics. Oxford: Oxford University Press Calvo, Guillermo, Alejandro Izquierdo and Ernesto Talvi (2006) Phoenix Miracles in Emerging Markets: recovering without credit from systemic financial crisis NBER WP 12101, Cambridge, MA: Cspedes, Luis, Roberto Chang and Andrs Velasco (2003), IS-LM-BP in the Pampas, IMF Staff Papers 50(Special issue), 143156. Diamond(1982) Luangaram, P (0 Miller, Fronti, Zhang(2006) EER paper CSGR version Stiglitz, Joseph (2006) Making globalisation work, Stiglitz, Joseph and Bruce Greenwald (2003) Towards a New Paradigm in Monetary Economics. Cambridge,UK:Cambridge University Press Temin Peter(1976) Did Monetary Forces Cause the Great Depression? New York: Norton Williamson, John (1999) Implications Phoenix Miracle Average growth, g = 0.05 Productivity fall = 0.1 Ouput fall, 0.05 Depreciation say 0.1 Savings rate say 1/3 sA   = gc implies A/3  0.1 = 0.05 Hence A =3x0.15 = 0.45= gc No fall in productivity, b =-0.05=-canonical growth. i.e. 5% off capacity Period 0 Growth 0/1Period 1Growth 1/2Period 2Q100105Zero105Y105C67 63.363.3S=sQ33105/335I=sQ-bK105/3  bK=35-11.7 =23.3, so b=0.05Where K100/0.45=222233Zero233DK =sQ  K-bK33 -10/0.45 =33-22=11105/3-(0.1+0.05)233 =35  23.3 - 11.7 =0DK/K= sA--b.45/3-.1=0.05 Canonical growth 0.15-0.10-0.05 = 0 Assuming b=0.05 and gamma =0.10 Period 0 GrowthPeriod 1GrowthPeriod 2Q100105check103.5Y-5%95approx+8.9%C67 63.365.7S=sQ3395/3=31.737.8I=sQ-bK95/3 bK=31.7 -11.7 = 20, where b=0.05Where K100/0.45=222233- 1 %230DK =sQ  K-bK33 -10/0.45 =33-22=1195/3-(0.10+0.05)233 =31.7  23.3-11.7 = -3.3 DK/K= sA--b.45/3-.10=0.050.90x0.45/3-0.10-0.05 = 0.135-0.150 =-0.015 more or less KEYNESIAN CASE Average growth, g = 0.05 Ouput fall, 0.05 Gap = 0.1 Depreciation say 0.1 Savings rate say 1/3 sA   = g implies A/3  0.1 = 0.05 Hence A =3x0.15 = 0.45 No demand fall is as above With demand fall of say (4/3bK) = 29.3! using bK of 22 as above So assume demand falls by double 20=7.5x 4/3, and use bK of 15, i.e. b= 0.068 ?really??? Or assume demand falls by 10=7.5x 4/3, use bK of 7.5, i.e. b= 0.034 only 3.5 % off Check this: why the big fall in capacity? Higher bK and higher sQ? Period 1 growthExpectedGrowthOutputQ100105105103.5107Y95958.9%12.2C67 63.3????63.3????73S=sQ33105/3105/336I=sQ-bK105/3 bK =35 -7.5= 27.5, so b=0.034105/3 bK=35 -15= 20, so b=0.068Where K100/0.45=222233233-1.0%230237DK =sQ  K-bK33 -10/0.45 =33-22=11105/3-(0.1+0.034)23.3 =35  23.3-7.5= 4.2 105/3-(0.1+0.068)233 =35  23.3-15=-3.3 DK/K= sA--b.45/3-.10=0.050.15-0.10-0.034 .= =0.016 more or less 0.15-0.10-0.068 .= -0.018 more or less Assume for simplicity that consumption depends both on capacity output and on current income, specifically C= (1-s)Q - s(Q-Y) = Q -sY So Y = C +I = Q-sY + sQ-bK =(1-s)Q sY -bk So (1-s)Y = (1-s)Q-bk So Y=Q-bk/(1-s) Should give gradual adjustment. Greater demand fall and smaller cut to capacity: Even more true if we use full multiplier of. So assume demand falls by 10=3.3x 3, i.e use bK of 3.3 not 22 as before so b= 0.015 Period 1 growthExpectedGrowthOutputQ100105108.5Y-5%9513.7%C67 63.373S=s()*9:;QRSadrstvG J k l m }v}}vvv h:hh h:hhxW h:hhFU h:hhv h:hhC h:hh:7 h:hhg h:hh` h:hhAf h:hhH\= hH\=5hH\=CJaJhH\=CJH*aJ hH\=H*hH\=h1`h:hh ^5h:hhAf5h:hhFU5h:hh1`5+RSt         gdgdhgd; dhgdH\= $dha$gdH\=      ; = > \ s  C S r  " # A B C     ϺȬȳȳ֥֥hf);hg h:hhgh> h:hh3Y h:hhXkH h:hhy h:hhC h:hhv h:hh:7 h:hh h:hhFU h:hhv h:hhxW h:hh h:hh`4 h:hh$7   G O j:FW{CEGHIkmȻݴݭϟτ}vv h:hhXkH h:hhAf h:hhA1jh:hh~1 0JU h:hhm{ h:hhs< h:hhK> h:hhMGI h:hh~1 jh:hhxW0JU h:hhxW h:hhZ h:hh` h:hhX( h:hhY: h:hh1` h:hhg h:hhf);. FG    !%& $dha$gd/Z $dha$gd~1  hgd7dhgdXkH hgd| hdhgd ^ hdhgd; dhgd; 6y{/Ny(;KLMVW 'OPbs}ݺֳ֬ϳϳϳȚȚ h:hhZh> h:hh(S h:hhd h:hhI h:hhm{ h:hh~. h:hh (0 h:hh| h:hh; h:hhwpE h:hhs< h:hhMGI h:hhXkH h:hhO<<<6<W#   !wx9:Ⱥsll h:hh/Zjh:hh~1 0JUjh:hh 0JU h:hhi h:hh~1 h:hhi5h:hh~1 5h:hh/Z5 h:hh h:hhAf h:hhXkH h:hh q h:hhK> h:hh ^ h:hh (0 h:hh| h:hh(S h:hhs< h:hhZ):;#%&7kl "#+08>EZ]hv / 8 9 D E P Q U V Z p { !!!멢멢 h:hh ( h:hhrh:hh2"6h:hh>)V6 h:hh>)V h:hh7( h:hhq8 h:hh~1 h:hh tl6 h:hh tl h:hh[ h:hh/Z h:hh |h@!!$$ &(((u*v*w*x*z*{******* $dha$gd $dha$gdEc$a$gd hgd7 $dha$gd,M $^a$gdA` $dha$gd~1 !!.!5!K!L![!\!b!{!|!}!!!!!""""J"K""""""""##$#8#=#D#a#i#q#}####### $$$_$$$$$$˽Ķį˶ҨҨҨҽқҨҨ h:hh,M h:hhA` h:hh; jh:hho0JU h:hh'> h:hh h:hho h:hhw h:hh L h:hh[ h:hhd= h:hhp^Eh$7b h:hh7( h:hh[ h:hhr h:hh |h4$$$$$%%5%6%E%F%V%%%%%%&&& &.&1&6&''2((((((((((())$)%)5)B)C)D)O)^)d)h)x)))̾ų h:hh~1 h$7b h:hhK>h:hhA`CJaJh>CJaJh:hhauCJaJh:hhA4CJaJ h:hh; h:hhau h:hhA`h:hhau6h:hhA`6h:hhA46 h:hhA4 h:hh,M h:hhS3))))))))))))))***G*H*r*t*u*x*{*****************~sl h:hh=,Ijh:hhWU jh:hheUmHnHu h:hhejh:hheU h:hhEc h$7b5\h:hh5\ h:hhAf h:hh h:hh~1 h:hh$7b h$7bh$7b h:hhK> h:hh; h:hh2"hl*x h:hh,Mh$7b'***,,,-..b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/ $dha$gdO< $dha$gd**++d+e+++,,,),,,m,n,o,,,,,,,,,,--...s.|.~......*/,/2/C/P/a/b/s/w////뒊h:hh~\G5h:hh/Z5 h:hhv h:hh |h h:hhK> h:hhO<h:hhO<5 h:hhl*xhjh:hh0JU h:hh^Ph$7b h:hhz h:hhohe h:hh h:hh2" h:hh 1/////////000&0?0A0N0O0P0Y0b0c0h0p0|00000000001122 2223&3'3(3)3+3333(4)40434=4D4P4^4̺̘̘̾̾ h:hhk h:hht$Qh:hh}6 h:hhXhXjhXUh< h:hh~\G h:hhAF) h:hh< h:hh3Y h:hhE h:hh} h:hhX'Ih:hhO<5h:hh3Y58s///0000)3*3+3v4w4x4z4{4666788 $dha$gdk$a$gdN[X $dha$gdE $dha$gdX $dha$gd< $dha$gdX$a$gd/Z^4b4r4s4t4u4v4x4y4{44444444444(5?5@5X5a5x5|55666"626M6T6V6W6y66666667(7*787X7Y7Z7c7t7u7777777ƛƢƢƢƿhl*x h:hh h  h:hhPdhe h:hh!" h:hh~\G h:hhAF)hN[XjhUh:hhN[X5hk h:hht$Qjh:hhk0JU h:hhk h:hhd;777788888:::::F:;;;;<<0<1<2<3<<<<<<<ǿvn\#jY H h:hh/ZCJUVaJh:hhA5jh:hh/ZEHU#jY H h:hh/ZCJUVaJjh:hh/ZUh>h:h h:hh/Zh:hh/Z5h:hhd95h:hhO<5h:hhX'I5 h:hh h  h:hhX'IjhA0JU h:hhAF) h:hh!" h:hh~\G8:::F:G:;;<<g<h<<<<<<=I=J===E>F>>$a$gdA dh`gd/Zdhgd/Z$a$gd/Z $dha$gd <<<<<<<<<===== =========== > >>>E>n>>>௢pc\\ h:hhQ6jh:hh/ZEHU#j6[ H h:hh/ZCJUVaJh:hjh:hh:hEHUjNI h:hCJUVaJjh:hh/ZEHU#jX H h:hh/ZCJUVaJjh:hh/ZEHU#jX H h:hh/ZCJUVaJ h:hh/Zjh:hh/ZUjvh:hh/ZEHU >>>??&?'?@AB0CCC D DEEEEEEEEEuJ $dha$gd $dha$gddhgddhgd6/dhgd/Z>>>>>>>????&?'?,?-?I?O?P?@J@R@@@@@A A"A$A&AAB$BCB_B}piah:hh{6 h:hhsj}h:hh{EHU#jNI h:hh{CJUVaJjh:hh6/Uh:h h:hh{h:hhQ66h> h:hh6/ h:hhWJh:hhWJ5 h:hhhZ}j,h:hhhZ}EHU#jRNI h:hhhZ}CJUVaJ h:hhQ6jh:hhQ6U"_B`BwBxByBzB|BBBBBBBBBBB/C0C:CCCCCCCCCCCCCDDǙǎ|oǠǎ]#jNI h:hhhZ}CJUVaJjӹh:hhEHU#jNI h:hhCJUVaJjh:hhU h:hhhZ} h:hhsjhAhAEHUjPI hACJUVaJhN?k h:hhj-h:hh{EHU#j0sNI h:hh{CJUVaJ h:hh{jh:hh{U!DD D D D!D$D0DIDNDlDmDnDDDDDDDDDDDDDDDDDEEE E%EEEFEREeEfEgE}Eƻʩʻʊ}vnvvgvg h:hhh:hhCDH* h:hhsjh:hhCDEHU#jNI h:hhCDCJUVaJjZh:hhCDEHU#j$NI h:hhCDCJUVaJjh:hhCDUh> h:hhCDhAh%+ h:hh h:hhhZ}jh:hhUjWh:hhhZ}EHU(}EEEEEEEEEEEEEEEEEEEEFFfFFFFFFF$G%GGHHHLǼ߸sfb]b hk4H*hk4jh:hh0JUj=h:hhEHU#j^ H h:hhCJUVaJjh:hhUh:hh6hAh&h%7t h:hh>hN?kjh:hhWU jh:hhk4UmHnHu h:hhk4jh:hhk4Uh h:hhh%+ h:hh%+"uJvJLLLLLLLLLMMMQ7QbQxQQT8TTdhgdC)1gdIFEdhgdk+ dh`gdU_udhgd Z hgd hdhgd:h $dha$gdL LvLyL~LLLLLLLLLLLLLLLLLLLMMMMMMMN(NNNNPO~OOIPJPaPbPɷɨѨѣ՞qjNI hC)1CJUVaJjh:hhC)1Uh:hhC)16 h:hhC)1hC)1 h(5 hU_u5 h:hh ZjhWUjhgUmHnHuhgjhgUh Zh:hh5 h:h5h:hhk+5h:h h:hh:h h:hhh>)bPcPdPfPgPlPmPPPPQtRR TTT4T5T6T7T8TJTnToTTTTTTTTTTTT.Z/Zܷܦܔ|wsksch:hh(5h)'sh)'s5h)'s h(5 h:hh2uh2ujDh:hhC)1EHU#jNI h:hhC)1CJUVaJh>jh:hhC)1EHU#jNI h:hhC)1CJUVaJhIFE hIFE5 hvL5 hC)15 h:hhC)1hC)1jh:hhC)1UjOh:hhC)1EHU$TTTTTTWWjYkY-Z.Z/Z:Z;Z^@^P^Q^R^S^`^w^x^^^^^^^^^d_o______M```aaqaraaaabbcccc c!cccch:hh^P6 h:hh^P h:hh/Zh:hhAf5h:hhS5h:hh S5 h:hhEc h:hhz h:hhI h:hho0 h:hh=,Ih> h:hhMt h:hh;{ h:hh h:hh S5ccccc d dKdLdMdRdidrddddd e e`epeeeeeeeeeeeeffffCfofffffffffgֲ֤֝֕֎~zh/Zh:hhz6h:hh=,I6 h:hh=,Ih:hhAf6 h:hhz h:hh/Z h:hh |h h:hh Lh:hh tl6 h:hh tl h:hh*# h:hh^P h:hhAfh|T h|Th07 h076h|Th076h07 h:hh07.fffffg%g&g?gWghg}gg.hPhhh$i&i8i:iHiPibixi$Ifgd gl gd( hgd7 hgd=,Igg%ghhhhhhh"ijjjjjjkkNkZkkkGlKlVl\ldlflmlslllllllnnnn`odooqqqqqqqqqqqrrrrrWr\r`rerfrgrr㳮 hC5 h95h9h95hChRwhCh+dh(5 hvL5 hARh(h9Uh(5h9 h(H*hARh(H*h5h(5h(Bxiiiii&kd$$Iflֈ  q 8 t044 layt9U$Ifgd gl iiiii$Ifgd gl iiiii9&&&$Ifgd gl kd$$Iflֈ  q 8 t044 layt9Uiiiii&kd$$Iflֈ  q 8 t044 layt9U$Ifgd gl iiiiiiii$Ifgd gl iiijj9&&&$Ifgd gl kdk$$Iflֈ  q 8 t044 layt9Ujjjjj&kdS$$Iflֈ  q 8 t044 layt9U$Ifgd gl j*j,j.jxjzj|j$Ifgd gl |j~jjjj9&&&$Ifgd gl kd;$$Iflֈ  q 8 t044 layt9Ujjjjj&kd#$$Iflֈ  q 8 t044 layt9U$Ifgd gl jjjkkbkdkfk$Ifgd gl fkhkkkk9&&&$Ifgd gl kd $$Iflֈ  q 8 t044 layt9Ukkkkk$Ifgd gl kkkl944gd(kd$$Iflֈ  q 8 t044 layt9Ull#l$l+l4l;lDl$Ifgd gl DlElGlKlLl9&&&$Ifgd gl kd$$Iflֈ 4q$ 8 t044 layt9ULlPlVl\l]l&kd$$Iflֈ 4q$ 8 t044 layt9U$Ifgd gl ]l_l`ldlmlsltl$Ifgd gl tlulwlzl{l9&&&$Ifgd gl kd$$Iflֈ 4q$ 8 t044 layt9U{l|llll$Ifgd gl lllll9&&&$Ifgd gl kd$$Iflֈ 4q$ 8 t044 layt9Ulllll&kd{$$Iflֈ 4q$ 8 t044 layt9U$Ifgd gl lllllll$Ifgd gl lllll9&&&$Ifgd gl kdc$$Iflֈ 4q$ 8 t044 layt9Ulllll&kdK$$Iflֈ 4q$ 8 t044 layt9U$Ifgd gl ln.nBnDnnnn$Ifgd gl nnnnn9&&&$Ifgd gl kd3$$Iflֈ 4q$ 8 t044 layt9UnVoXoZo\o&kd$$Iflֈ 4q$ 8 t044 layt9U$Ifgd gl \o^o`obodooooop>pjpppprqrfrgrrrrrrrrr$Ifgd gl gdCgd(rrrrssssssuWvnvovpvtvvvxvvvvvvvv%wDwFwwwwwwx,,,,,,,,,h-n---.....//4/5/k/////ù䧣hih~1 jh~1 0JUhxWjhxW0JU h:hhCUh9Uh2u5h9UhC5h2uh(h>9959C7@7B7J7L7N7v7x7z7|7~7777777777777777777777䮣䮣hk+hk4CJaJ1hk+hk4B*CJOJQJ^JaJmH phsH hghk4CJaJ7hghk46B*CJOJQJ]^JaJmH phsH hk4+hk4B*CJ OJQJ^JaJ mH phsH h/Zh%7thk4CJaJh%7thk4CJH*aJ,77777XXX"X$X&X(X*X,X.X0X2X4X6X8X:XXDXFXHXXXZXzXgd>gd|Tf Potential gc gc Supply Demand recovers ln Y ln Q Fall in demand bA/s Potential supply path without crisis b 2 1 0 time Loss of Potential 78XXXXXX X"X>X@XBXDXHXVXXXZXxXzX|XXXXXXXXXXXɾ}}y}y}y}y]7hghg6B*CJOJQJ]^JaJmH phsH hU_uhg+hgB*CJ OJQJ^JaJ mH phsH hghgCJH*aJhghgCJaJh&hgh>CJH*aJhgh>CJaJh/Zhk4+hk4B*CJ OJQJ^JaJ mH phsH U+hk+B*CJ OJQJ^JaJ mH phsH zX|XXXXXXXXXX YYYYYYY Y$Y&Y0Y2Y4Y6YZY\Y^Y`Ygd>XXXXXX Y YYYYYYYYYY Y"Y$Y&Y.Y0Y2Y4Y6YXYZY^YYZhh hqfqfb^\^XhI;UhiU6h||hghgCJaJ1hghgB*CJOJQJ^JaJmH phsH 7hghg6B*CJOJQJ]^JaJmH phsH hg+hgB*CJ OJQJ^JaJ mH phsH hU_u hghghghgCJaJ%hg6B*CJ]aJmH phsH +hghg6B*CJ]aJmH phsH !`YbYdYfYhYjYlYnYpYrYtYvYxYzY|Y~YYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYZhhhhhhh hh h h h h hhhhhhhhhhhhhhhhhhh h!hgd( h!h h:hhC,1h. A!"#$% DdX  D  3 @@"?Dd" jpN  C *Aphoenix2"+|w9f,O5@=+|w9f,Oа`.Vzx՜ t';S QA#$[Z5U,!HKQ!| +001"&V=3)f?Ͱw&\moB?wiD_GhڼnѴV[]5-$jZViՑE? /FҴDW]Pm6Xh}Ije6EQ4hL $ӚR\տ?oii^3ܕxh ajic EI}]YUkmc$ˡ}Ly9eIzA\@ݛ{w@ tMَ3Нt+tyh= @+6tۡ{FG8RrltGXy^o6I]&K..Wհ/|fdI yɵOfTLv*<1ǰjO c[m&Vx1qCNҤPR]IwDs㟿@o]tiF2GlOMXْu=J;nBo]bY{Ӯ6LZ/Xvyk H3ǰjî7XfQڑêZĚؾ1|^D m?X,t8WEFOCwit_UROIZh-fY~]S<}-PE%-m=1Z"Gץm]m[ FF*/ygO #mv/l$tr2s#SԷW'&ǃ%= ;+׆wK{ ?O2??$( ?ϟH'' w-Y/+OÂ?IA0'?<σOZ[ ~?A*f)k8x\,'1WnI~/~k'^6JE8xc/sT8~7ݒ^{-]T_=N ?%9e $0_!n1$ϧZW|4%=I q\-_AG<_'gH-qlH%s龜?d7IN%qwHI,U'+}?-$Ϫ>S /O/KU]?S{"φfKV?U9j2 1,׀I  AMyV _"Ճ參Ou)K%0yVퟘ2L7|0[ zp+/U/O3?wMP#d_D~%o}yI_{X>Կ\}P a(~I~}Tn 7 /VXA7K4E_~3Q/C)ߏui2/C/(ʭ&YIBukZFWke-a-cWm\rӚK׶lsX]CQږqh-aVúkᑸ\Q/%eqOFt>yF>]Mo>i& ?@-ZTgO$'w'+Ϩ.(ID'ZW5SOuwkTgg QB #O$IQeFƅ࿆;7=Yw1?&-WfM$CLt G݈;Y'  yn-/yNEWǁ$2-T]:'$ohEZ$i??-ôjﵚ<6H˽sϛFhB#>vcR ;,V&A Ҽ@FWB%-}HWG-^Ѽc3QnZ&a'9c-ceט>I~E}s8U+FI4׃?#U'+Q2Y?mFV mo_UZs$dU o?oIwy1%-Z%=¯jN/KexRUM$9MC!I?i9ԸadEn*?B?4~0j'$#{.ym::G]s,_%Pr;x_9bI~&U=ǖXF=sT=ϳ9u>O~xO?ϣ9Ԩ|]'yͱ]Z Bs_N<Σ7< <. %\I<3Ah*GC;aK 蜞6W/2y\v ]#T;-2Lhh2u HCBҮ|hX֛<6Pla"gV{"]U8{<3cIEvJw:ڴՋ]kHa7X{shcgs*03,3-0=l6B`xm5CT4EW٘E(:2م52&ݬك)s3v<~j\v)u/?_?E?/=G+ H_$/I sָ/s#G-}T'{9 sU*Iy XUq_+_ j UWvK_׃~9aM֭ 8|Dd!N  C *Aphoenix1"ڈKJ;^5@=KJ;p!HS3|xy|E YA$dm`;Iw'Wd :۸,: (+rƕup{GDž]@T@tYޫ';u*O]w}ʓo0 Ùs Xpܹm=&9wܹ$Kuy@@`N0Je6)} _-te 92ޙ !9sIB69w<<2ibϳ`Xzq)|;{wuE|*{>=/wt̾Wfξ;'m1܅}vסvuw? C?PEK%WJ{4S3O/A]~ ohlO~=GB0{1mN%<پU]{O?'*K=>;_ݙ'Et0??{_9]if{S{=Io!Yg~f{ mNFPm{dT{/-BsҜnyN5WJl4ŻmuO.hwˇ Gh.hX:0xRnJ|n] A n_(9Lif{߇}H|LJPPIW$A~nh>EnϟDNWA.?$A_I\Nxg.hwi ht:hd-.8Vf{tC{J?!3gNg3?[@[b8KhxJfx&xRnIa3i_9Osd_[RYd'Z-Cuw\+_-W6տ-έm?ŻEDRAN3%d_[~f{43wt_INTŒI(x4BnODoћ!d_[~f{(o=Y#llw/o [i:ɧ9<ݲEN~[soΥ7ݲ@.t-p݋heK%Żţ.jwru~r+ν{K5ۻ, tZq~nh~"oUڿV:$5I_[~f{~t+CӜ%Fg$/-[-XtLN8IwI_[~f{|toWuhzʱ@rq-+;huA[1GtKNo(5]o^]mx|4Enq gh?n)~nbt_LN戔|T7x01yi4ۻ%GKա_5o%o9 %ŻEh~Di^ncѝ^tɫђ~nѼ 2ѝN{Nn 7Jw̓,w/o2p#ho_ZRLJɵ~n|s4ۻe&>T1H&9JŻEh4f{xhE3tv+s%e~nu{ս$CRy&Z[-cYd'[) /-T[03itT8j$2Uir7G}3(;Hz8ŻEhnP[-Yw^ӿK %={.jw˃o5,RgMSNK>hwVse4ת?+rru5".>ڀSgnL@KmNm^L =Wu?3>?U\W[6o |qeIAޏ 9/_tZw.Q_JmoVl,[Em@gϝm7L|]+\surAH:^@\nOWkgJ}?@f~럫Jt_swqPӽݕ_q*s?]\LN_q*s?]-;>no]%~7]u;Νm?oxJ.=/_?+]/?7s{du;~?[Ec럫Jt_ "SOc6 ?z]?Vo^\nOWkg]p wLT玶_VgwQXFo?WEӕh>|./>3ˢm/!` O=32%s=8&J&w%-a|1q(pbX< P2vsSJhQu t1C1K+]VI\4B+$$qIXq ꈩW4@43>ʵ<ŵ a19oXfʈ)cMě39LaB M&tq3gnr+SEUWwz̺;Ъ 0 KP<[Ƹ vjSVQ*%ɳ|K T1_݁zZdLJXEPXde`.a.~A>5@ds%P@'8,Q,o1I b4 zY,bls1w9oyG09C;,$:=C9Ԟ'CKܩ<ȇB(1^~_\\D+~?ҳ&0]>#HFj5R}~ٟN_|eu[OQC{Sn^*d66YLȀt$dݧ`,yԕ YeYN,[MќFjwlwm?ɐ8+e^,KXjK }};e/b9~yC$B2> >){]j)U\-IDHxc_X#Cs!z xN?qq >9 "tE1ۢ\hx<0bcA/tK;0#~DxEGus5(. %\{{{%;.v1o{uj<p<IIj[7p8|C7 ͌"5ݰvпhNvm'r N.؍ | ? G;BQeQeݣ="kE-_? Vjn"NV s#cMЌf46Sc3:M,h:k%_wA\Ip$_qA9T@%XƘ*ꮦjb휹 P ` eĔ&M쯉3˄L,C̙[* 6TQG՝;0{h0T@%Vlݬ}< 3g`yk#.W?Wkx%C ߎ੎A<׷ Y'MI!7= ;I!\,مF؍ vvvƶ3g;sc;w$\Jv+E J3v | 0}G;JQe;Gp=[::k4LQI"2! >#˱F5黑"Cn۲n!^< r\{԰Z?eCa:s:ioYxZJ Ȅ,G>\j܋ۃ ;a~pk;l;~l;l;l;>Ѱ߶?-a<p#;:JΣ?:GX~}:B y!r#>GX0k 쇽vѿS֔s2Ɍ3o{=7Jq= SP x# '>D9 ZC ъ(ꍄ2#7; f;SҋqAofN/`'xۉNvn'vrm' v$n{_3.G>aM8*"!70އ䴥ʳ4><;:񮔠 k20L\2IAc]1^|QGFPGA.$,iW1&3VA#]iT E*4椑#]M<736@H`dODC"K$Gɛ4s j_C֠kGw$S u!B!M^`b }xJPG5>| 5A=4@ZYge0/H%&T b ț4gwZwZv {fp*P.ɒXgf쿛T ˜ +6UQ{5uWg]{hm*,+2&晘o"&tX2 ξ*ʳ WWE ՝;0{h0ׂrET~64hlQJP< ,3͇?;Y&VseZvPu o|kAO:ym#]l ̒4B+ss=?L~ ft d[K4@͌"5dx4'~<Ɠc&wh2\'`̄f;Ǽ̟M٬; f u}'i#}<'t-'TI5u) SaL'~:1yB+y#]HO5:~n{7(f3|?T@Xyf ajg릜 +6UQ{5uWg]{hm*,+#Xkb&kB e-`ˡ:pm8{y|-ȡ_~ +_{7W>T??xYZ Ȅ,9cdm#ed앑^ aa0dA&C㩴ZzA3tS?4:ּn`Y&2 4НTt9M?y03+wɵK#4@N򪤖9n^׼/ PKzE4(&g_įe3x\Ȼ/,<7Ex^ULkoao!B`3kZt7~js%u誇hYx3lI#4@=_x-$~@;0{.a%VTT@W4W3%x^ܗ9_"n¯=_z k[˜}ϐop=<'|6&x^WU7koaoa-Bȿu^W:t^qڻ_Ou 6mhS6P P.Dɷ625w`:GT=&S EPX\0=E{/B(l"^dG5>5-xG7 Y!3R6Rj!. ,ȄtOk-A_oT 5>e|^eͬ&2 4T"ݩNCЕ Yɳ|/K>ɛt>G4>#t>Y#Ų䈥ЗsKݱ^qqg<$@"$S{OޤK܂|6Hz7\ByY:aZI_!Q ?lWP|ÏA9) >d"}?Aq ݌:Na&g$Msss:I-sܸ2'nꈩW4@4?pcC1198yO~SNpi9\ML31 O:خY?m:خx O$xcxcccc(q|q '8 ?6pp{y 8?;ú.tut]Cq؎pgϡϢ fvhSp~)!d7g sv98 .|pwVC-h9pé-}Dp"! 5p>H,#?}ps~qb{y'$&^ˏpzc#s8z'z$?3.h#M='uQsaI{ΝHM%aQ-:ɗj|!_GA8* H Y1Cɗ=  P/)oz0J.#]0 Jt c\PʜRri&hflq-"(|ȅ0sCJkW*a0. ę0| y,k!Y2Z'wQW<ɟG~A.kF.kYZyPŒ&j4BC"݁Vufz,be*PN 3컙4sx;$P`cv⪨}fٿ*Ι PPFLk"<{g" -&4Xf3@9TlۡuwU݁;`FrET~$ y0^B\ %|EPO9u&; )1y.y\I>">3$)YPX.地8=ᬆ̌ڈDzX[G|.sQԚ')fb⋙+(!W yXgmyhj<O"IWd@&dA6CHMFW59! Y <1Jew1>>5ђ{:s:1T٧;i!2!ٞb򕰪ot:ZYz(DsF1^._K/tǠ?欢^|Qǽ:o:Hyoz/-:nsފu3~-\73~qs/#u,yAg`7]Wp܍g#i܍pJG͇+{3n E\609A0o9k.yo{ E{n{y1>Ǻ(ZBo:OM)/񦷐f nbX&ָ ͂YֹmpPG67]󦃔 K3ݭtM/Hgf͜3pswqA9T@%XƘ*ꮦj휹 P` eqq&M3w&rbB eh.`ˡ̇pc+⺫-̽]p7C̻>rG{YֻuDx;ttYoz'y]A=qk͚wֿyp LMH$H?qq >9X1J;B/QDC}xuΞ!@s$DAΧ3QIgnMDC㽈͜޴} 7?9ar!Ed[a6oD0jc؏0&;݊h70އ>열>?XѰG/۰'Σ*ڦDzƛ&켶[67~{ļ{{onߦʗ<6|*>C~tS{+03̾7)U- ʡ* 6UQ{5uWg]{hm*,2L쯉y&gB ]&46/Jl;UUQCuw3z谠\Q`Ƹ v49[;:m+R{%T@9 fƣ|ߗZ꘨'򦆷;|>s9?>Ŀ㷡߇+0mӠk|ƗncCu?dBӗxSѝT4Y Ȅ,Ϧl}KRK5xwo[&> xi^Ƿ{CG|,bK}ߗc?}e?qqϹ&@"$l`?q)̽ЗY4>$#և)³ճޢM$$ӟ̸8-,fc$ھ԰ѱѱǷɠcٰ&:tcʏű!׸պ;ЪLccOl (WXqbfD@8pb̏ W9#:=Y7,ګ=fC;mP<[X1eĚ_{f"|&eB_Z8k3gnr+3^E\՝;0{h0łrET~64ک͎N;u*B%T@$|bC:&j IsyȀLȢ?!6ۈn#bDs1"ȦN_|eX я5H֌h֍a]71TRѝ4t ]% yC|B:FƏIa?ppps x$B8b+e^,cXK }ߗc?}e?qqi<$@"$g`?1أ^D U,<~,#XձS1QQO)aa;uK%Pvl[oYWZ6|ZpF06H_ǭhfF1B\ y[߂maQ݌Ɛgp旒|@.C=g7IuvW0 H#(%WIP U ;6IBN*I %hjgf*M7h@IL>7EP(|$]Cϵ| bI9!7vjӕ`XO >MdBdSj2RF4oLB,ȄtOk-\A. fnNĺͬ&2 4T7ݩNCЕ YIZ#uLSNPO^vѭ`+\W3[nG yg,9b/5Rw,DzG?8Hd00֟ħ0 F~'< Ml$A"$@<DZq84ǡ='!dI5>@x]-,F{=dCjm5 I/KoT_Ÿl[W):y ߧOa=^7 +$5lqXu,a\8s'hfFZ x@? $Cx݇Bx 専?ɼyb'ɻB2/ u^NxG0CkKu^x?iRexxx57PN"Y&13̾9O3g`,XCTlىjfϪٿ*Ι PPFL{l"<{g" -&4Xf3pP VE}v"NYwZf4,)WT@%X鷡Nvt$_Lb3Oy#"X#/ W?xExEx˅R1.aEPG.CJP eE-"1I. rF9{arIgPs /%n8Ù7 #o \y;7[eB QB(|yԔGmnwSP(f\PB^ KiOㅗ2O K%+ȿЗy7Md1㋉[Lb.&I q% y;_W<)IWd@&dA6CHMFtk|"yG0>BGɹhzYA.U}/5r]κ+$s:1TtߩNCЕ YEy|qy:/<_4^xQ#?! !%!>sGa!,/ rXA =9"X܉^w*/| ?AaW ZΚ`)?a!<$! Of@x~j?>Ox\$/A4D GPW8y84GBDk|:#ZEyFP!DA4(z1ދMA `x\A  I|Oy*1.xh'XFa>aM8w* so/aq}/|>ʰ;;J%K%g_;WvK>O~0 ~(h! NTy'5|A/V s#MЌf46Sc3:M,h:k%~'ث`+5ŧzϋ=TT\f/Uv.ةc¸*`cN\uWOW5W>9sXBʈ)cMěgbL2ń&:-0s(J~;TQG՝;0{hXS.ATl&gNw m/};^͟܁z@װOR,ُ܏g'N7{`7tԱ]r< PQ % #Oٯ_)83/b(ȧ?r%>y9]rВ\ JQJ|iw Ow&;0{R[B JXj,AEPOd;|?rXݗ)Z_+2  !왑=3r~FtlDH$K_Kr~_֠?:\հRג}:s:ivK]%N }wRWjvx/^3Oi?~| /O_.x~b39gYw:vvwRuWjn셯ɹ{?kg}5eݰ/s)kJDH9d 7EhQuz(/:(z7\By>s )DxE=&Y7kyM2 e,}9Tg";fhUw``F= +6mhi^+yX7g&;Phށ`A@204_rYh$@\A01A>[(湄aĕRpΞ gJaw ^ P1 C0/9!oq(zs9< y.aWJ ށ<CCz|(B(b/Ac K8tPo@x󒙟L$HPx֬ЎE叶h<4T/ 2!lbgF~nDFtA͆ij%K,v6kU3ӳS*Y*\2u[YZ`,jT@%X1n#Ngg1\wU޳^v)e]Bӗ'* 8xa ?;y31\39YgMek11txgQ;I)޳(aOPP1\t_ԒPEPL1%ĕ?_qnS; atJE_61C>o#Z#Z{ 9ij >) SYwɄ H?4t;ݩNCЕ Ȅ, ehU5c`a?w|/$A"$@<}8bIe50s'ds+LiNg r t֙zSYZtLNݿ+^KVJi$H?8ֈc8ơ;!HDHd7g 7~wFhhW~v^X ݮ 01_-朢83Ao|^s罈O:d//(6I&x NPRZ|՟OF4Įf,r ,e < YgcȳpO֠OgX˳Xzծgɱ52||^[xՒfiFhzZj|ӒA5䨡jj2 E-ķ0BNV=G<7IꡎZk#Yc$1͵h:ꩇh乑&ƛk&ggQ;VULYTլ_*ƞ P 7[&Yu,l$fPV1ngfnugaם-96y PYw#?`- 4%Pɳ~6}vug1Q=c)뗢cz%y1A!@>IV㓞 ePt%Gg"͡P!|(B(ƇWgNϢDwT SgQZ%)& cdz罶"Dz峧2闑kh5<1 Kc< Y,oknYwɄ H?MugtȀLȒ܀ǻm(! "l '>pOD; %%1OgD+7"8`s6mS: 3ƒ ~+CKSwp~aab3~SpI^8:;ĜC=DC<=={c㌟\9?19Nϊ66Kjx;8<,i&h:eL0ؑ9B{8'$k9%uL(n!:P$)m 5ncYD (a0q?̼#pߑ8#{Bw۔:?1G00^ P0fk"0D&D&={ CXF"՝]w8E|5?)? 'Ò Jqqv}wԑMy:SŒ}?Mp!0n:TD{)IWd@&dA6C8#c|췑wd!scw?Ͼ@/g5IWpVb͟%CYdBI~bݓp|LJ Ȅ,1' 8#xB۔Wv w}?ZwJg}(XQՔy9CO)ڟP#ծ~կz8TzC1tutqUvvY12sR$DhG0FhM"c$XDhmI.> M`lFP 8O:PQs 6֑< G[tg\E(渣7FRs|̱'FJv ;o C?-??1-_ XBǺ)uC ttH3F. []X ݰ;0v )nv'B;Ѯѯ=]By~k1^Z<2>чy΅] Y)dPiH4?SYTlOT69He.ҘE:sِCy.>ⱫEߴܳ津~127θ)ސA:̰)崡Ȧ<>41RoT[bս1RGJHrtؤH`wwt.E(HsHuv1R#Ҥ-+8ڳ>IOT :@2\G"vIv5q4"OuvHZxb.hvӴcNE5Sq>xVv6m5ɌӁ:BgGbȮV~Pcu}b>Z2g--- ͡4&ЈO4>DCCo ^?yfNAcֳ 4fuC vtt#;yeb947 11v7&h ͠9PC},bi47Ӻ8Wz覩(|QDB]@U6׆"@]4b}"Ru M;ԅHQ_ScDh縚R}S Bݴ$NH^с8vLQk{M&Q͞: QPo@Wh>O+[ɳ+"2·RmfL N4Ǧb$^9A3F(MvEW^t+O1}1%@L8{' S8*G4g:}g0L4=0nIbi~b+y'T0 &M{Ү+>~c>[>ONY?lƙ͘{:D{:L"V+Tb6F8qfLl.R.EKЛ ` L|"L~&~cMi!Ŋ$.I0bũgaD@lt/qR& SȠ.]3o"L" S`*L`:̄ڔ,̢lƙ͘{Dw:6L"V+T|hcĊcc̈́-%+E 4`"4!shKܐ{=cďֽ'K/JGOQ|JaagάAgĚ Mb @l9QB+zZL p/ZӧŌSx::R߁vɴw3}Oi}=1f顄qy~q535Rb9Y֚wf"+%N-!f0 MɴL8>;q&3b>\c1?z|hcı3М %薠[I6x;;Y@\юD|K}b==C3Һb>Z2g-C-K-Sh͠)4Kw/qH: &J: τ(,͢lƚ͘:Dw:vL"T6F;qfL,R.E4g 4`"4M4c],Cr~֧bo1I>D2Q)ĄSͦMvGa uY/EE(uq"2·16bWLl!^RafxF `4"^1bl 謧n=}w=g f6aE3bbv |g}_ch*vh4!@7dd{cuIˇq >0ݰ zg֠3vwҼaH >6iRhB{EOM6YO6m$Nۄ/Ml-Ց{O'l쌭]wФhB1{+dNb,Zm RHؖI؟I_I_w"!8&X1oҌ54tw&@k㩏88l׼DmiG]"%>1dV~ri{Z1-{-C-K-Sh͠)4\ј646Ë^"n ko跋3:Bg'z݆[߶o)6;Swh*  (ߨi ͠9PC},bi5!琯y;;DB]Cym"h[j3n^l+[i^XB|`=#&MEQ"!J EMMri點13݀5h_qjxISB[){ESk{M&Qy͞`oՁH>WS߀v |Yf#ReE-flUpsLPYTtt8GVڝ+m:T0qE `4壩/m}qE\"H{X<}<\ 6Hzx5 {AuwqĮx^)4qxf# wLKbʋpybw.] S\$)_3F>>_>W>~O^X;E)DM\ o`|0$lı5Jy#)Sޏ~hQOs .E47R7vh?(dke-"<`Ces4ߝ'λɀtڤ6MLe-R#HT"HT!Hcә Ȅ,Ȧ,\aWZxY 7HJƺJ ! 25W{%5!LȂޜgCu9˥}h+-1Yr3ÕptvAW뤹ّ#މ\@WeX+tYS1A\r(:37] SH] =-.^$U\yyƸxWq8to݄MIcAә$3OW3tBy'mzҾgH33l3tWz%v@<6/ - 5 ̇"im!ZSO}M2𿶜Nߘy0؈@J ݚI6Кx;8neWkh Vcc;9L }}n\[#\+%{%{%/bؓ͡4弉И6h_4bCh!hZ6fM@Sh1Rׂ6-VmU}.>7LbuE+qbOAShByM 4fМMU341smg3ϛ&V3؈{eAiύ0 i҈5+||Ƿ||ǯ|)` k` &|L sb[cĊy\Oy?/ (d= Çs Fnl-bZ1y>ؓ 9̹" SȀti ̄,Ȇ}! iOg@&dA6P\}}\\/ka.(c3V/zCdj G?r7dBԧi7ٔPC\ڇ&V+q+%PHt.ЙN|⣾}8t.UK_EWogMcdBNY:iЍ lTdjzZd@1"T71R/q4Gс>}7)tEGM.u}/C!eO~P=7D XvK5b~K'~r-$b_{K—$J$|M$=Hd~A[H֔Skk,n_xp+E3 -DM֔ky5hf˅Ŋ9~xKX7YZC[Q*1W*1W*1W*1W*q7"~g4cDCC|hlO4s6Ƭgh bo<}+>w&k1ºŢ81h͠)4 h4fcscloM)4PC},bi\Mì\JP@W".ԡ.BMVy:]"5(֧*"["X`ܤSHMm藮G> ,2 6Tڤ>Xtg-MdBW>GkK-|5&sP9͞: QPo|5}h>_]3J 4߈szja1JXAXN8t)CĥK`11"a!, #a\QD"+Z,0ͥ\e9xx>]$,|Kg]F]+qW(f^3Jl3߈.u>#NK8q.\bʹĖs1i iWH??1-_ XB֬f-e̵D{9,"f^3J9ψ3 Xͫ`5cFc5vBo%+`9v,Ű0OS,`(f^3J|6߈.ĭ `>1<ŹlȂL!t|x,%rX hg 2Z\kXJtc*b~1LJyF̼1V0JXŸ}WU/S& zC69.Y|>w̹o=`.ئżb;C'^+)Wtb](}C/%1#[MlcFZxg-Zb57Víe3xbhh!}w!,bR|1 fn+a]I6x;X8nEklk Vqb,DqRTF\XjqBX%K3xg.q\,!ЇH8Ji1bb>(b}".륈♵|Zg3xaA͘N6Exހ}{5#`$є+hWD{X}M݌1%nهo oq6]3 UCw}0A44EP~xf?~=_;Y٤hm rhG}"%>1db+kq싑ֵ؊hdd/Ze?@shM ЈvhMhGG4>4?yf> A Ե]KVmUw-a]V0vF0ZTD=o򽚦 C 1.g}NV>Zg|Nu8A8D8L&B#hAU9weOTEPsni0/ Q}ئx[߄7 ,^=nnX{wԢ}-yނ5.2;Κ[k25KmT{DB)k|srzV_Ok/\4c<+ww6=9Í0C39Ss#sТT( D{}hűh:甛<(̥lnysE&dh (%84ۢlh/==r.E[,|qy7RZ<$, nJC{4h*dr˅U휫5myf%+[=$4 DZr8xs>i|̇Gcv4UnR2ʆND&˛ru?q‡? VO7;?hu$)嵜7jh?khܬhMߕ85?XT04՜h {.MC9TnRig/±^} ͆v)h*z !ӱQӎ->Ҽ(l1گhDSِ*/*RcW5[h2տ[MeCh^=co;[ׅ]h5G3dnr/+.;}^hoF{KʆޢVWW5ok^qE;of>dhwE3 Mo=rD{&jxs!h2Ch*o{G8(^IC;I4 D5;N[8N '9?O ml(e;Y }DO[C4@/ x4 .ўW<5s./u|iP^^R/']ihMeў(*_S!Lv\d0q*y]=j.v 4o@SphTnrEppEuxG5-h09MeM=aw cNpǕX\8jh]n 4 7 GMWiM-knpEP;@7 l&yHI_(+qEoGM~^Fvz}lEzO\R-4Ѽv8]Xhnݜ4 #EˤEs!ƕhaZh k =:ʆѢhZhp ׸~&NVv;CмJ}f#m\^Zst`ȑh-4ԴDvm;64T6(GC=\#YhU>vkߍ6Ѯ2BE誱9R,…B7λhO5Wsk*n. .{qt㲧Pnj}ކ._lK?gL(rpF-wFN4 {\GYX1О9bgT6+\ -opcbw@a_GfC4T6L\^inr5x`}G[ihߋahjaff}%(Ѯ-ԤTe hN͚^2٠#~ş?oq>.)40Ѵp;8k>Ħh)>ʶ>(<ý5p}횇4wW砩l'c\Z,wӬ HjFhV +5~ށ\4 DfZ5!<@ybnh*ʆŢoeR qT<ʵ"K&sn{2T6<$i_< _skWOo}_:57MeD..BW3$uB9gAC{=KT6,6?<$my׵>\_1ڛ ? MeJюotye}EcF5+T6]hjkf𚫎f&&T6+;\^vjZ lawhWOQѾhpI+h.0xU>h7#h-ɿ-ކka{S;~h}^ch*։Ƕ.'m \A1>Yhv(Olh}2y]1w8z),@oh4נ$ʆ_.oZ|$|m;4A 4)n;2a?- /k ϣ_lxZ72٤ܵ6yVnŗ?/4O]ih/Fh*~+*\YO[*CzW{h.hve MXjm6,+l5h0{4 ω*ja!89wV,]4W5( v Ϣlh/լ0YdQbŃsl_>4  3[Pzl |OsP_}ڍ h Me =ְO*&rtmq]ƚlGkhOBy4 E{Y[ n5jmk(M !V=k b4DE{^ZU TFs-bp&N3vmvwE.Ϡ-Aֶ"ϖhоG/4ףо h*^L[RI:!T['롗ЁfhAs ʆ֥¤]j:R,:$ۺsTt ЎG;h*^v&+Z,mVgyjvS{|Mek–QaRY Zڲ4l9>4dW4Qh7 Men`S'BCM'QŠHn5Q}&h0oDs;9횶~&445l7Z8-V84x= @|xXCh**a*TZ ќ=Me" /5ÅG[bT?4#y=2@8 \4wl'ڧTx8^c*X|lʼn|8nq0.@wYhESh~{ɇwX @=pw{4 ؼk\U GE6Ҷ!p,1V%yn=hfV<7o;mE_ws[X\7-4yn?"m_Oۏyn='n?'Գ/=8β;<3[q>˲'幹Mϲ;yԭ*ϲ<~yn$msynsKs qck<ϭsE7͏yn%ϭo s<7\4f<_6HSu .}>“< ܼ9nmyn"ϭVst钙T֜ի^Vvk؈~{:+*:ϵY݋hٱM9~UU{eN)iͫ=*l鿚ZvE}ej|jF>_{MiiZ^BXLu3ٳ#w=ה37,E=dxr,Cބ}w8cEW{?kQ#' {^|=]ւKOm7SS{U{G>wr*t)Ņr&Ly-}dCrhH/ty^GJk#c >>2;>mGں~G\㋼W1o~,QtI_9}|=tiLKYs?HM:i|#=\.mϻ'{ ' ƅVv'k;<Ŭ(ֹ(a+{#pz~[nu~ՑH6#~cHc7_WY#e{\s}M_gau}kҀ=פ{5ܞs/=aA{.F6=7Jl9uOPϵ_˸|kkV\ d칅ƚL\bj=kr712?Ϻ3`-zZlXIk-&YAYwvV=vEԀȓ r->X;Ĩ` b A_)Wڙo?X 폰:z2_gziCdi={{_gp-6 ,`= LkꀵX) r-^_^6֢zZ6=ZA-4"s?\ud-Sr}?|.ƚyoǺ5Az5Q^{}$u_^׉_&iSkN|jgIeZ &:MAIwo"ƚtⳚ#&eMv&E&Eـ5yS75)>4!֤>>Pj>=)>5W#<p;ŶZU̺5ܷ// =nW_ ~uz|5b o 7p\溙ĕ;LԵSzpqeKIvx*sya9;֢^uF{lF*}?/?`Od/5{,K^ROo-[d/5 Z"{i?7` E*jzh+?N?wO\ゴ~~,F yΓa\J?YOr^q,w r#?syY3S_=|~mזui2}u=IT5]g ԛ&.=J__O( ?$甿9gBܰ͝?[Nx5fU _UlU9:,}GS7w/귚zE\"*ﱷ} Wa+rsNǾbse4Ƈ;8pmbz Csin\C8r.vq>$ӴUYM*\o1]}F잧!b9v)롧 לFKq ]I>p |H'@W` F&eMjMz2M|Aڨ;ߎ4cڃՂ_[ ߴ48%k;09-`f9׏}(doC_Cy6mzӶ7kvcQvGh\rޏ@ho)3q\:}!_s04 سo\c=r1.C 2tWu> zZ\⿪D %_w/=῿`t?#{b eÍnf]nSp0yn05\ ~*Og}~[q0@Q0znk7Sd|L;7"a [6n 8NQF2SF_1R׌PQq/ca Q);w18 >9YѴK<oeEhSQӚؤ(BS18U&)dI?sf~ٻd,]]>hO{fʜvmΜp?8 ,LҜ=7QsF8-pb' N 3踙؊`Ӵ=)l< Q>|b5uet~:{7ƒl+J?JCQ/n>;sa~ p:E/O}Kq\S1,bfa#wqqBs, 48 gf1Xh@sVRż6y0a0ifa.Ӝ_ilRBw?oVԿY-ٹ63Y., K5g )(XQ>ͱջ+(Xþ?TrP)9\2Y𲵍FV9´~],H9䏲>W?ի{SO?ʥ9ݳ0?'<=Bӷx4_[.nsN{\sϢqhA46(NpS~2ڕѾ~eWcp>z#^9>jm;,~3s))ts\8Xч~cLųCq >i 78#,EϏ'6ϔVt.* Wuu+NU\[~{v;^PDd H@b  c $  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~6  ,  !"#$%&('*)+-./012345g789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefhklmnopqrstuvwxyz{|}~Root EntryZ F"/ .Data {WordDocumentY.nObjectPool\/"/_1210079725F//Ole CompObjiObjInfo  #&'(),/014789:;<=>ADEFGHKNOPQTWXYZ[\]`cdefgjmnopqruxyz{~ FMathType 5.0 Equation MathType EFEquation.DSMT49qn  DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  Q==AKEquation Native _1210079710 F//Ole CompObj i FMathType 5.0 Equation MathType EFEquation.DSMT49qn DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  DK==I"-dObjInfo Equation Native  _1210079466F//Ole K FMathType 5.0 Equation MathType EFEquation.DSMT49qn DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_CompObjiObjInfoEquation Native _1210079484F//A  D FMathType 5.0 Equation MathType EFEquation.DSMT49qn DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_Ole CompObjiObjInfoEquation Native A  d FMathType 5.0 Equation MathType EFEquation.DSMT49q4DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E__1229894545@1F//Ole !CompObj"iObjInfo$Equation Native %_1210080054 "F//Ole *CompObj +iA  I==sQ"-bK FMathType 5.0 Equation MathType EFEquation.DSMT49qn DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_ObjInfo!-Equation Native ._1229892178'$F//Ole 2APAPAE%B_AC_A %!AHA_D_E_E_A  bK FMathType 5.0 Equation MathType EFEquation.DSMT49qLDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_CompObj#%3iObjInfo&5Equation Native 6_1229890790T6)F//APAPAE%B_AC_A %!AHA_D_E_E_A   DQQ== DKK==sA"-d Canonicalendogenousgrowthrate 8a"g cOle ?CompObj(*@iObjInfo+BEquation Native Cy FMathType 5.0 Equation MathType EFEquation.DSMT49q]4DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  g== DQQ== DKK==sA"-d"-b==g c "-b FMathType 5.0 Equation MathType EFEquation.DSMT49q DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E__1229878064.F//Ole ICompObj-/JiObjInfo0LEquation Native M_1230018298O3F//Ole RCompObj24SiA  Q==1"-g()AK FMathType 5.0 Equation MathType EFEquation.DSMT49q DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_ObjInfo5UEquation Native V_1229891352;8F//Ole ^APAPAE%B_AC_A %!AHA_D_E_E_A   DKK==s1"-g()A"-d"-b==g c "-b"-gsA==g c "-b"-g(g c ++d) FMathType 5.0 Equation MathTyCompObj79_iObjInfo:aEquation Native bL_1229892091=F//pe EFEquation.DSMT49q04DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A   DQQ==sA"-d"-g==g c "-gOle hCompObj<>iiObjInfo?kEquation Native l FMathType 5.0 Equation MathType EFEquation.DSMT49q‡0LDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A   DQQ==s1"-g()A"-d"-b++gA==g c "-b"-sgA++g FMathType 5.0 Equation MathType EFEquation.DSMT49qLDSMT5WinAllBasicCodePages_1229892388EBF//Ole sCompObjACtiObjInfoDvEquation Native w3_1229892746GF//Ole |CompObjFH}iTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  b=="-(sA"-d)=="-g c FMathType 5.0 Equation MathType EFEquation.DSMT49qObjInfoIEquation Native  _1210080895LF//Ole LDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  g=="-2g c FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjKMiObjInfoNEquation Native _1229902231QF//n DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  b FMathType 5.0 Equation MathType EFEquation.DSMT49qOle CompObjPRiObjInfoSEquation Native >" DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A  Y==1"-g()Q==(1"-g)AK FMathType 5.0 Equation MathTy_1228663825J,VF//Ole CompObjUWiObjInfoXpe EFEquation.DSMT49q2 DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/G_APAPAE%B_AC_A %!AHA_D_E_E_A   DQQ== DKK==sA"-d CanonicalenEquation Native N1TablejSummaryInformation([DocumentSummaryInformation8hdogenousgrowthrate 8"-b Balancesheeteffect(Fisher) 8"-sgA Effectofrecession(Keynes) 8A? ?3"`?2*G>[u Iwڙ5`!G>[u Iw@" xuRAKAf6[M`݀TA%ILv[W5D֜^ z(zAf݄pay0Dbɴ9c!bh9zt>s9EIKF=wte8aPrԪeu ⼧h?ȴ yʛʵߐdT.TugۨV5[۰֖@RTjr[wdPr{ͺ^3SSo3˦QL3j~{$Ptt&B6 dU?o6%3Pd0NzH@"zsXe!h{TL=3_CgZ6]n5$Bq/oSﹻ, a Dd b  c $A? ?3"`?2WNJc@)}35`!+NJc@)}  HxuRMkQ=M6"bg ؅4 ZBJ0).ǩmJH]. J.\TED\II>w=w Q@ 0\* d⣫t>͉i\L1btELd0glg=N7DPvz[CY2h^ܥ_ 5ZnXwvolF{e885M^RJk%ʔ AkvS) K&Yc}ql.hS9pPB.-0lj_TQkׁԝƎl:ƽ|Vn 5cB jJNC$"Q|8Hb\2͑|Ӷ* V;[%oj3yKƫ>oy]g1b-HbETB/n^9k&yZgΓ:#sz3R(]t{n Ra恉g4}OI_V]Dd 0  # A2ng(@BE[ǟ5`!ng(@BE[*`!xMQM/Q=MGۘi!Œ+؍aSi +mĂD$~5?@jH}/s3w ( (C3{mMA ߸0?c묗 h`@'{2B~TIi],݂ >lǢFFa>zOQ\oɌ]U(&\2 7ڏ *yw-WEیܔVQ3[5VܺZ*8ED+UrnE<"Hf~ .sfeMsfɶR+k75mt-F֨q f@ xM‡츝giaFc0N4`4j4 (qVl먣Sޮֈh,ݨ 4#8)"9eҋ"MlyނDd 0  # A2rZEc̝~I5`!rZEc̝~*`!HxMQN@}s!TBqADAq%A! %_4H|5| B"XyogμeR5%Spg *1yO>UN>cވ;ѕ#ӡY0).Pl}ћiەwPVXE$ [)xz ϙ|ZQDd t tb   c $A ? ?3"`? 2$Rn9ШSwp5`!o$Rn9ШS   q =xڵUkQvӤY4 XAՊh !)/6Y@[7)7P xxҳR"(4W?lQhf~fc D|\8x0erۯ!jM~9 в<>&Z )'S3S 8'ƅ@3/Zj:mhHkX/@5̔LZ|Q)JVUGjeEib3HDEf^ܻ́0 jj>#JU~aV|-/fjӥ lg/{id}_l CeW`[0Ic'&ƝY{kaI8%rgyݳWLou|R?ݼ{g{I^d Prt+s1pH2uϼ-R>Ӳʐٴٳՠ`'芾ziԳOٱٳ٨vX٨ʢ>ۗ-᫱<&-h?[ւq|" `H qͲw{wGHl8dJ$dntqS! ^рĉ JBs<3[)Cm0l&d8\DĤ;8e3ցC-[,98"3eݽw$ݤl4H; s 22S*+E=:pHE9?}nqDd b   c $A ? ?3"`? 2(]bduӝ~5`!(]bduӝ~L !dxڵTKkQ>I)y jх,IP-I Kc<$ƸB(]k.JuQ7E覈KIǍyEPf|s7?G A Md̲}:[Hۡq`Ok4>p0k]tj&;܀E}/@aQ#6Gw:8ATfGF-ϔZSBӝ[o헾Mn&:'d$" f殱Unr#Wl7Y.97f 9,)5Ւ$?omTcaa%ݩRGcE^$|\6Suq zcL-tVGiFH$._$^-n)l/..1sSZB^Ԓքkn>կ5Z,(Fe35<&O=֗gd}1z&g=0edل.~˨k iP8YN>QrU"[qSfsYc񠗦@Ttw;~%AI D4)疱*S+~ TTliU H%r~⹣Jn+ UDd b   c $A ? ?3"`? 2 AS R@NN{q5`!s AS R@NNX %hAxڕSkAf6m0?b7m aծ14% ^ADo" "EAAxCk&)kK鰳7{7 ~@:U!9V˶)_檿s ΘAdЎ>+Q//qa3 k;|Tԋ7J_ٺ%/>T/l>ny1FT.*w}{P7>es bh9,9wYQg̚zT4Ϳ\uH19ء8zuYmz'zL'=ܧFvUT<8T.hx!:}cۆ L*{ /j¤QL̴3+n-ÒOXNfYDؓʪɻղ|mDǯZ9xFK[ǃujLd ZꫦBXB׬Vģt`ĨP vPges?8׃ Dbs#Ɉx =V4(k=tTGuP^E Qvj+5lWV3KЛ͌&Rlb 7WҲ4w Dd t b  c $A ? ?3"`? 2(gNm*K.:5`!(gNm*K.:  qdpxڵTKkQ>ܙس'-;)"ݤ&&d<$وc!B\@袋QW?ЅY(8n#.T~s73F A &2"F'!G.T0%%A`kTy=۾a`/hkJeNxF(_h @m?r0%]W5x6՟stN)Ix]Įsbwerq4TUOJ ^zݴ-(LlUq”DJ)aqOoDd b  c $A? ?3"`?2Mw5Ӯk(b!)5`!!w5Ӯk(b! ,dxڵkQǶZDvSM-i"x0MڸDAVك~ =xxTVĢQbQhPD}7f~o&aP C@yC$bє%]Q`GI\EqMs4sM 修tgsq%!w6ـ-fyߌGM V:—9(5og3@pe8Mʻ0w3Dp@ -kf#Seba`PV`N$Th-1:-s|`1Ԣ`Ĩh6,FYy11w8iT_FyZaWYUu4ui lzܦ)X7*S+E\6:gVg |x/Қw͗z-fLf cp/|:DJX<x_KxL"jʼnLT۷'h|N-ThtiP)*vEHri7 | 0Km` >Q iPEHufZ+yazJI9 3dbq!:`R3-mliW{u%xC(LhpB|CC#;mbt V  9PHrpQ2e8)ӮUsT^z)lF5+^֤,{,;1qd|ǭU4|l-^V2ռU)&%3/|㺆{/Qݨ !)ՒwcVsdM9U_&_U/v'sk-^CUAF邔*G\t Fu-1uJoy0-)j !yH a[A8w:5i[Mqx^~z?j0їDd h0  # A2r+F8Ja5 WNN5`!F+F8Ja5 WN@|xڕSoQ-tj&85$z#qK^Dh H /x^&/N8A>z^)г$'l4OdWb&%3!'gjQ?m s[m`M*e7*?X8zDdl  XD  3 @@"?Dd @0  # A2ܷX`p uxi5`!ܷX`p uxi*R xMQJA=36 +Aɚm"D,׍ȋ$ӥ?`eaMV{k%שׂ8eϜ̽g) Ē)8cbhz\9x S0H&BPHCHYlbiWePq.[_n;80j/is2]ʞdK~Q+5!U~ D&]v%PXjxyN[v 횶i7r*A.T9^>8KeIj!}akm}a4褑v"'L;5PHӯծ~t"U6M0 %h i̐hO鵔o <CwwTYE@xF:q斈ڗ0~|t/Hn7rx{Ddl T  D  3 @@"?qDd b  c $A? ?3"`?2J8%>SzB65`!J8%>SzB6@h]xڕTkAfv74mEb ì4X()IbIDb.RxE<(xC =x"RfI4m)Ǜo޼o. Mqc b(tsVe[c촃 򎟇kY]*FΘFgߋ iǠA"[/e@5!k+|Tۥ`x7}wOgq[&0ySE솽\E-%P ӥ6.zB^-/ߌY(K`8H$ n+ W\=עk =W/6CrwrUW!GjC?bEp4\暼@98:87` 'S`f~(&XiLfsf^_UdUbYʉ/ZBqf_Y#iT Xq՘Ȑ1z5G}nNYg2%.C0Rq r%%F(JjO#a.? y~uqxLhHx1`Nj8f%H+Umq6xق:BWdDd t b  c $A ? ?3"`?2(gNm*K.:5`!(gNm*K.:  qdpxڵTKkQ>ܙس'-;)"ݤ&&d<$وc!B\@袋QW?ЅY(8n#.T~s73F A &2"F'!G.T0%%A`kTy=۾a`/hkJeNxF(_h @m?r0%]W5x6՟stN)Ix]Įsbwerq4TUOJ ^zݴ-(LlUq”DJ)aqOoDd b  c $A? ?3"`?2Mw5Ӯk(b!)5`!!w5Ӯk(b! ,dxڵkQǶZDvSM-i"x0MڸDAVك~ =xxTVĢQbQhPD}7f~o&aP C@yC$bє%]Q`GI\EqMs4sM 修tgsq%!w6ـ-fyߌGM V:—9(5og3@pe8Mʻ0w3Dp@ -kf#Seba`PV`N$Th-1:-s|`1Ԣ`Ĩh6,FYy11w8iT_FyZaWYUu4ui lzܦ)X7*S+E\6:gVg |x/Қw͗z-fLf cp/|:DJX<x_KxL"jʼnLT۷'h|N-ThtiP)*vEHri7 | 0Km` >Q iPEHufZ+yazJI9 !Ey)iW27tٜ.yX wIN'9)@@l ;wLiUxU'_:8 |6=IXw ޕiعX0v([~[$Im쨜\?ʊ/+•HI,xvP.%y9 ᜕-Y9{go#:,N_ Oo0?m%'r!b?gn<$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555 585#vq#v#v#v #v8#v:V l t05q555 585yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh5q555$ 558#vq#v#v#v$ #v#v8:V l t05q555$ 558yt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh555T5T555#v#v#vT#v:V l t0555T5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9U$$If!vh5555$ 55#v#v#v#v$ #v:V l t05555$ 5ayt9UOh+'0 0 @L l x  @Crises in emerging markets:Fisher, Keynes and Phoenix miracles ECONOMICSNormalecsax2Microsoft Office Word@@4^6@/@/FW՜.+,08 hp  WARWICK UNIVERSITY4!f ?Crises in emerging markets:Fisher, Keynes and Phoenix miracles Title  F'Microsoft Office Word 97-2003 Document MSWordDocWord.Document.89q^ 2 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~_HmH nH sH tH @`@ NormalCJ_HaJmH sH tH DA`D Default Paragraph FontRi@R  Table Normal4 l4a (k (No List >@> ` Footnote TextCJaJ@&@@ `Footnote ReferenceH*4 @4 /ZFooter  9r .)@!. /Z Page Numberj@3j ( Table Grid7:V0PK![Content_Types].xmlj0Eжr(΢Iw},-j4 wP-t#bΙ{UTU^hd}㨫)*1P' ^W0)T9<l#$yi};~@(Hu* Dנz/0ǰ $ X3aZ,D0j~3߶b~i>3\`?/[G\!-Rk.sԻ..a濭?PK!֧6 _rels/.relsj0 }Q%v/C/}(h"O = C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xml M @}w7c(EbˮCAǠҟ7՛K Y, e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+& 8PK!Ptheme/theme/theme1.xmlYOo6w toc'vuر-MniP@I}úama[إ4:lЯGRX^6؊>$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3ڗP 1Pm \\9Mؓ2aD];Yt\[x]}Wr|]g- eW )6-rCSj id DЇAΜIqbJ#x꺃 6k#ASh&ʌt(Q%p%m&]caSl=X\P1Mh9MVdDAaVB[݈fJíP|8 քAV^f Hn- "d>znNJ ة>b&2vKyϼD:,AGm\nziÙ.uχYC6OMf3or$5NHT[XF64T,ќM0E)`#5XY`פ;%1U٥m;R>QD DcpU'&LE/pm%]8firS4d 7y\`JnίI R3U~7+׸#m qBiDi*L69mY&iHE=(K&N!V.KeLDĕ{D vEꦚdeNƟe(MN9ߜR6&3(a/DUz<{ˊYȳV)9Z[4^n5!J?Q3eBoCM m<.vpIYfZY_p[=al-Y}Nc͙ŋ4vfavl'SA8|*u{-ߟ0%M07%<ҍPK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 +_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!Ptheme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] $r,/~=n NO :EOn58^adgjpr &'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~n6543210/     ] ! " # $ % 1&/-,+)('&%$#"! (789:;<=>?@ABCDEFG H!I"J#K$L%M&N'O(P)Q*R+S,T-U.V/W0X1Y2Z3[4\5]6^7_8`9a:b;c<d=e>f?g@hAiBjCkDlEmFnGoHpIqJrKsLtMuNvOwPxQyRzS{T|U}V~WXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~:EOn58^adgjpr &'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~nn $$$'  :!$)*/^47<>_BD}ELbP/Zv]cgr/4j67X h!h<>?ABCEFGIJLMOQRSTVWY[\^| *s/8>ܴTZ澱iiiiijjڰkkٱձٱlllǴƵlnnrrr"(   B C0DE<F$ )CPPP!^,.)0 0@     3 GNB  3 g 3  4 G  BCDE<F$bb,bbbb4b b @    c[(x   07v/ 7~ ! 66:N 6~ " 65 i  5~ # 64J Y 4~ $ 63O p 3~ % 62 p 2~ & 61 1~ ' 60 * 0x ( 0/$ x /~ ) 6.j . * # NB3CDE8F { oNQTN\Q_Wa^^eXh yr o)  !?B?93+)JTNFCACJ/}5{<}?B?9U747=DKNlolf_XU*ou}y4%.('%$!$* " kcacjpwzungdadk ;OV]`#c*`1Z4RURLE> ;  KUOHEBEKh0I6G>I@PCV@]:`~~xrjh! 8;82+$!,pw~z6/(%"%, v s v! |$ !  l N / - / 6 < C Fv dp gh df ^c Xf Pl N Q W _ a d a [   !                    M z \ Y \ b i p sW P I F C F }M z 41 8 ? B E "B ); , J M J D > 6 4               - r x     | 7 1 ) ' $ ' -  `B? B"H%O"VY w z w q j c `mw1q3j0g*d$gmRY`cec\ nlou{NF(%(.5<?X]R`J]HWEPHINF39@CFC= .ssTyRU[ai}k92+(%}(u.s, !$'$"%CFC=6/,o        y s k h f h o  T Z~ a d g d ^        Y : 8 ; A G O Q p r o i c [ Y O        Y) S, L) I# F I O  4 : B E G E >         g d g m t { ~        0? t { ! ' - 5 7 :V 4X ,U )O 'I *A 0?    " % ( %         p        z t m j g j p k UM [J cM eS hZ ea _d    | u n k %        < > ; 5 / ' % Q        [ T M J H J Q  6y <w Cz F I F @        Q 3 0 3 9 @ G J h k h b [ T Q 1 v |      ;" 5$ ." + ( + 1    $ ' ) &         r~ ` ] ` f l t v | u n k h k r~ 7 V ] d g j& g- `0 N Q N H A : 7      R\VNLILRd7 F= CE FG LJ RG ZA \{}{tngd         4 7 4 . '  3!w!~!!!!!!=!6!/!,!)!,!3!!"r"o%"r("x+"("!"!!!!!!!s"J",")","2"8"@"B}"aw"co"am"Zj"Tm"Ms"J#X#^#e#h#k#h#b#### # #  #####$$ $$$#######S$v$X$U$X$^$e$l$o^$W$P$M$J$M$yS$v$08%?%F%I%L%I%&B%($G$I$G$@$:$3$0%%%%%%%%%%%%%%%4&y&&&&&&&>&8&0&.&+&.&4&&\'>';&'>)'D,'K)'R#'U&s&v&s&m&f&_&\t''''''' '~'-x'/q'-n'&k' n't'(Y(_(g(j(l(j(c((((( (((((k)h)k )q )w ))(((((((U)B)$)!)$)+)1)8);_)YY)\Q)YN)SL)LO)EU)B):*@*G*J*M*J*D*)))) )))***************5+oz+Q+N+Q+W+]+e+g?+9+2+/+,+y/+r5+o+(,  ,(, +,-,+,$,!+?+B+?+9+2+++(v,,,,,,,,,y,r,o,m,o,v,-[-}a-zh-}k-n-k-e- ---- ----U-7.4.7 .=.C .K.N-l-o-l-e-_-X-UV........`.%Z.(S.%P.M.P.V..;/A/I/L/N/K/E//......//c/`/c/j/p/w/z///////70;{0000#0)0104A0R:0U30R00K.0E00>70;01"1)1,1/1,1&10 00 0000w111111111{1s1q1n1q1w12g\2Ib2Fj2Il2Po2Vl2]f2`!2~22~2x2r2j2g2!23 3 3 3 33282;28212+2$2!@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `/ a ~ + 6-K -~ , 6,M ,~ - 6+ D! +x / 0* *z 1 3 (1C"? @ \  ( 2  BCDE<F$bb,bbbb4b b @    #"  ? . l  3  "0?`  c $X99?l NB  3 =2  # BTCDELF8{ X.TX5TX:T^:Td:Tk5Tq.Tq qqkd^X X TTd T T@    `@`P " NB  3 1 3 HB  # 1 3 GHB  # 13 GHB  # 1* NB B 3 1 NB  3 g >   B C0DE<F$ )CPPP!^,.)0 0@     3 GNB  3 g 3  4 GNB  3 13     BCDE<F$bb,bbbb4b b @    m>x  0 ~  6:N ~  6 i  ~  6J Y ~  6O p ~  6 p ~  6  ~  6  x  0q %  ~  6Ej   # NB3CDE8F { oNQTN\Q_Wa^^eXh yr o)  !?B?93+)JTNFCACJ/}5{<}?B?9U747=DKNlolf_XU*ou}y4%.('%$!$* " kcacjpwzungdadk ;OV]`#c*`1Z4RURLE> ;  KUOHEBEKh0I6G>I@PCV@]:`~~xrjh! 8;82+$!,pw~z6/(%"%, v s v! |$ !  l N / - / 6 < C Fv dp gh df ^c Xf Pl N Q W _ a d a [   !                    M z \ Y \ b i p sW P I F C F }M z 41 8 ? B E "B ); , J M J D > 6 4               - r x     | 7 1 ) ' $ ' -  `B? B"H%O"VY w z w q j c `mw1q3j0g*d$gmRY`cec\ nlou{NF(%(.5<?X]R`J]HWEPHINF39@CFC= .ssTyRU[ai}k92+(%}(u.s, !$'$"%CFC=6/,o        y s k h f h o  T Z~ a d g d ^        Y : 8 ; A G O Q p r o i c [ Y O        Y) S, L) I# F I O  4 : B E G E >         g d g m t { ~        0? t { ! ' - 5 7 :V 4X ,U )O 'I *A 0?    " % ( %         p        z t m j g j p k UM [J cM eS hZ ea _d    | u n k %        < > ; 5 / ' % Q        [ T M J H J Q  6y <w Cz F I F @        Q 3 0 3 9 @ G J h k h b [ T Q 1 v |      ;" 5$ ." + ( + 1    $ ' ) &         r~ ` ] ` f l t v | u n k h k r~ 7 V ] d g j& g- `0 N Q N H A : 7      R\VNLILRd7 F= CE FG LJ RG ZA \{}{tngd         4 7 4 . '  3!w!~!!!!!!=!6!/!,!)!,!3!!"r"o%"r("x+"("!"!!!!!!!s"J",")","2"8"@"B}"aw"co"am"Zj"Tm"Ms"J#X#^#e#h#k#h#b#### # #  #####$$ $$$#######S$v$X$U$X$^$e$l$o^$W$P$M$J$M$yS$v$08%?%F%I%L%I%&B%($G$I$G$@$:$3$0%%%%%%%%%%%%%%%4&y&&&&&&&>&8&0&.&+&.&4&&\'>';&'>)'D,'K)'R#'U&s&v&s&m&f&_&\t''''''' '~'-x'/q'-n'&k' n't'(Y(_(g(j(l(j(c((((( (((((k)h)k )q )w ))(((((((U)B)$)!)$)+)1)8);_)YY)\Q)YN)SL)LO)EU)B):*@*G*J*M*J*D*)))) )))***************5+oz+Q+N+Q+W+]+e+g?+9+2+/+,+y/+r5+o+(,  ,(, +,-,+,$,!+?+B+?+9+2+++(v,,,,,,,,,y,r,o,m,o,v,-[-}a-zh-}k-n-k-e- ---- ----U-7.4.7 .=.C .K.N-l-o-l-e-_-X-UV........`.%Z.(S.%P.M.P.V..;/A/I/L/N/K/E//......//c/`/c/j/p/w/z///////70;{0000#0)0104A0R:0U30R00K.0E00>70;01"1)1,1/1,1&10 00 0000w111111111{1s1q1n1q1w12g\2Ib2Fj2Il2Po2Vl2]f2`!2~22~2x2r2j2g2!23 3 3 3 33282;28212+2$2!@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `@    `/ a ~   6K ~   6 M  ~   6  !  ~   6 ~  ~   6  h   z  3  C"?x @   z ] 3 ]C"?  *   ^  BCDE<F$bb,bbbb4b b @    #"  ? B S  ?";Bn*X 4tl Ttl Tt5Zy7 Zx7 ZL'TZ 'TZUZl8 ZUZ]ZPZ 9: Z7: Zl; Z; Z<; Z|; Z; Z; Z,]Z$\ZsQZdXZ<9 ZlZTQ]ZXZI< Z,Z4]Z$XZQ8 Z\XZ]Z]Z< Z< ZT< ZT; ZdV< Z8 Z= Zs8 Z8 ZT8 Z; Z,= Z< Z < Z < Z < ZL= Z$W< ZdW< Zd< **vv'+? ,,c.p.p.00y277??0JQQcWkWkWrW4X4X?XXYYYYYZZMZMZf9g9gZiZin      !"#$%&'()*+,-./012341102L ,,k.x.x.00277??9JQQiWqW|W|W=XAXAXXYYYYYZZTZTZf;g;gcicin  !"#$%&'()*+,-./012349*urn:schemas-microsoft-com:office:smarttagsState85*urn:schemas-microsoft-com:office:smarttagsCityB'*urn:schemas-microsoft-com:office:smarttagscountry-region94*urn:schemas-microsoft-com:office:smarttagsplace=1*urn:schemas-microsoft-com:office:smarttags PlaceName=3*urn:schemas-microsoft-com:office:smarttags PlaceType H: 54341544444454'4'4''4'44545445451435445'1434544'44'""55*9E9;< B BDDUfffhh'i-i2i:ikkkkkkkkkkkkkkklllll1l9l>lDlllll%m'm+m-mmmmmmmmmnn{DDUfkkkkkkkkkkkklln3S H   s'x'.Z//002h444J555F6'7778u8*99::6;;;E<[?tAAFBZBBDQEAGTGpGGHo0%j;  ~1 (m{{I|Ecq8[*#6/>!!"2"AF)*k+~. (0C)1A4k4iU607b8d9Y:f);I;>;s<.=H\=K>'>.B`CCDEIFEGEp^EwpE~\GxGXkHX'I=,IMGIWJ#L,M^PwPt$Q(ST9U>)VJ>WN[XZ\0Zqz]% `1`$7b PbdPd3eQoeAfg g ag:h |hi8ibjN?k tl8qo qa#q)'s%7tMt2uU_ul*xy||hZ}\}}w/ZRwO<9vL SFU< ^I 7(AC/7 ( Q67d}Iee L`3Y'W&6<C%+ `4-6o@:Y|gyaur?vz Src[|TxWAvd=>ZkX($A1g5A`;{ZCowX ZUfWf@" Z7Z8X>X?8C8D8F8GZ[\]^_abdeklmnn@@@H@N@R@hj@np@t@,\@46X@UnknownG* Times New Roman5Symbol3. * ArialA BCambria Math"1h[[FW4FW44d!f!f2QHX ?1`2!xx>Crises in emerging markets:Fisher, Keynes and Phoenix miracles ECONOMICSecsaxCompObjy