22 Abstract This article shows that if, in a young democ- racy with weak institutions, one and the same party governs in virtually all upswings of the busi- ness cycle and promotes each time pro-cyclical fi scal policies, three serious negative effects emerge. The fi rst is the loss of fi scal policy; fi scal policy remains pro-cyclical during the downturn as well, deepening the recession and extending the period in which output stays below potential. The second effect is the loss of democracy; un- able to use fi scal policy to help exit the reces- sion and speed up economic growth, the parties governing during downturns compounded by the pro-cyclicality of fi scal policies are perceived by the public as impotent and are penalized ac- cordingly through a lower share of parliamenta- ry seats, until the party that governs exclusively during business cycle upturns fi nds itself without a real opposition. The third effect is the loss of conventional monetary policy, manifesting if in- terest rates and infl ation are low when recession sets in. Under these circumstances, lowering the monetary policy rate to zero might no longer suffi ce to stimulate the exit from recession and the quick return of output to its potential level, leaving central banks no option but to resort to unconventional monetary policies, such as quan- titative easing. Keywords: fi scal cynicism, business cycle, pro-cyclical fi scal policy, monetary policy, de- mocracy, political parties. WITH A PERMANENTLY PRO-CYCLICAL FISCAL POLICY, WE COULD LOSE DEMOCRACY AND MONETARY POLICY Lucian CROITORU Lucian CROITORU Senior Advisor to the Governor of the National Bank of Romania, Bucharest, Romania Tel.: 0040-746-102.625 E-mail: lucian.croitoru@bnro.ro; lcroitoru2003@yahoo.com Transylvanian Review of Administrative Sciences, No. 54 E/2018 pp. 22-37 DOI:10.24193/tras.54E.2 Published First Online: 2018/06/29 23 1. Introduction Over the past two decades, Romania’s fi scal policy has been mostly pro-cyclical, i.e. its fi scal impulse and output deviation from potential have been concomitantly either positive or negative. Pro-cyclical fi scal policies increase growth rates in the business cycle upswings and lower them even more to very negative levels in reces- sions. Politicians in Romania have not learned this lesson yet. In 2017, despite a GDP growth rate of about seven percent and an excess demand of about 2.9 percent of potential GDP, the government decided to administer the economy a positive fi scal impulse of about 1.4 percentage points. In this article I will focus on two issues. First, I will come up with two possible explanations for promoting a pro-cyclical fi scal policy in Romania during periods of swift economic growth, at rates above potential, partially fueled by the fi scal policy it- self. The need for such explanations arises because, unlike the monetary policy, which is partly insulated from political interference, the fi scal policy is formulated in an in- tricate political process. Neglecting this process may lead either to misunderstanding or underestimating long-term trends which, as I will show, erode democracy or to unrealistic conclusions about how to preserve a correct fi scal policy stance at all times. Then I will analyze the consequences of the expansionary fi scal policy separately for the ‘normal’ phases of the business cycle, when monetary policy rates are signifi - cantly positive, and for the ‘abnormal’ phases, when policy rates are virtually equal to zero. Experience so far has shown that, during normal business cycle phases, if the short-term Phillips curve (the link between infl ation and excess demand) is not fl att ened (i.e. excess demand generates relatively high levels of infl ation), a pro-cycli- cal fi scal policy during the expansion entails a massive rise in infl ationary pressures, which calls for higher interest rates. Then, when the economy inevitably enters the contraction phase, fi scal policy remains pro-cyclical, this time unintentionally, deep- ening recession and unemployment. In other words, the pro-cyclical fi scal policy in the expansionary phase of the normal cycle leads to the loss of fi scal policy during the recession and the exit from recession. I will show, however, that if the short-term Phillips curve is fl att ened (excess de- mand generates relatively low levels of infl ation), then it is likely for infl ationary pressures that lead to higher interest rates during the normal phase not to emerge anymore. In an environment of relatively low infl ation and interest rates, along with a pro-cyclical fi scal policy, it is highly likely that the entry into recession triggers an abnormal phase of the business cycle, when monetary policy rates hit the zero mark without the economy returning to potential, which would entail the loss not only of the fi scal policy, but of the conventional monetary policy as well. 2. A cognitive explanation of pro-cyclical policies in democracies with weak institutions Most economists and analysts have explained the pro-cyclical fi scal policy in Ro- mania during expansionary phases of the economy through populism and lack of 24 knowledge of the implications of a pro-cyclical fi scal policy. In other words, accord- ing to these authors’ explanation, the promoters of such a policy believe they are right in their actions, without realizing that they might add problems to infl ation manage- ment in the expansionary phase of the business cycle or that they defi nitely contrib- ute to higher unemployment thereafter. This explanation is similar to that suggested by Rogoff and Reinhart (2008) through the phrase ‘this time is diff erent’, which also implies the lack of knowledge, not necessarily of the negative eff ects, but of the un- avoidability of a crisis. However, part of this explanation – namely that regarding the obliviousness of the harmful eff ects or inevitability of a crisis – might not be correct. Economists have repeatedly explained that, in an economy such as Romania, a pro-cyclical fi scal pol- icy in the expansionary phase is almost doomed to remain pro-cyclical during the downturn as well, deepening unemployment. Figure 1 depicts the pro-cyclicality of the fi scal policies in Romania through positive (negative) fi scal impulses and positive (negative) output gaps simultaneously at work. It can be seen that pro-cyclical fi scal policies in expansionary phases have remained pro-cyclical during downturns and for quite a while afterwards as well. This brings up the question of why a government would want to implement pro-cyclical fi scal policies if it can anticipate their detri- mental eff ects on voters. In extremis, in a cognitive approach, like the one proposed herein, there can only be two answers: governments either use those detrimental ef- fects as a means to a political end or believe they can dodge such eff ects. I will explore the two possibilities one by one. -10 -8 -6 -4 -2 0 2 4 6 8 10 -4 -3 -2 -1 0 1 2 3 4 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 Pro-cyclical fiscal impulse (percentage points, lhs) Countercyclical fiscal impulse (percentage points, lhs) GDP dynamics (percent, rhs) Output gap (percent of potential GDP) Source: author’s creation based on NIS and AMECO data and own calculations Figure 1: Pro-cyclicality of fi scal policies in Romania determined based on the primary fi scal impulse and the output gap (forecasts for 2017) 25 2.1. The ‘fi scal cynicism’ assumption Capitalizing on the harmful eff ects can be an end only in a democracy with weak institutions. My explanation, which I call the ‘fi scal cynicism’ explanation, is that – in such a democracy – there may emerge a logic of alternance in power whereby the costs of a pro-cyclical fi scal policy will be borne by the current opposition (future ruling party), which will weaken and reduce it in favor of the current ruling party (future opposition). By virtue of this logic, any party coming to power during the up- swing of the business cycle would promote pro-cyclical fi scal policies. This explanation assumes there are political parties or coalitions that embark on a path of maximizing their political market share over the long term, and decide to govern, if they can, in all periods of economic expansion, which they amplify via pro-cyclical policies in order to weaken their political foes, and not to govern in any recessions. As I will show at the end of this section, in Romania we can identify a par- ty that has so far governed almost exclusively during upturns of the business cycle and promoted pro-cyclical fi scal policies each time. However, we have no grounds to claim that governing solely during upswings was or was not part of a strategy pursued by the said party or that it was guided or not by cynical or benevolent inten- tions. It is not the aim of this article to determine what intentions or strategies have guided or are guiding the pro-cyclical fi scal policies of political parties. The aim of the article is to identify the possible explanations for these pro-cyclical fi scal policies, and to infer the implications that arise if, for whatever reason, one and the same party governs during all business cycle upturns and promotes pro-cyclical fi scal policies. The objective of governing solely during periods of expansion (not at all in re- cessions) is rational from the perspective of a cognitive bias of the public: the sig- nifi cance in terms of advantages and disadvantages that an unanticipated gain has compared with an unanticipated loss of equal magnitude is stronger in case of the loss. As I have shown in another article (Croitoru, 2017), the fi rst economist to write about this cognitive bias was Knut Wicksell (1898, p. 3). In virtue of this cognitive bias, losses emerging in recession are perceived more strongly in absolute value than gains during upswings. Thus, if a party that governs in the expansionary phase of the business cycle and promotes pro-cyclical fi scal policies were to govern during the en- suing contractionary phase as well, its image would be seriously dented. By contrast, governing exclusively during periods of upturn, there is increased likelihood for the respective party to be associated by the public with the idea of success and welfare. In the theory of fi scal cynicism there are no guarantees about how long an ex- pansion can last. Expansions can be very brief. For instance, in the US, the shortest upturn after World War I lasted 12 months. In theory, as time goes by, the economy nears a potential recession which can set in anytime. Hence, for parties maximizing their market share by governing only in virtually all times of economic expansion, in the logic of fi scal cynicism there arises, on one hand, the need for fi scal policy to be pro-cyclical from the beginning and, on the other hand, the urge to step down from power immediately after the possible entry into recession. Moreover, it is sensible 26 that the pro-cyclicality of fi scal policy should heighten towards the end of the govern- ing period, which coincides in time with the approaching of parliamentary elections. Heightened pro-cyclicality may ensure the success in the forthcoming elections of the party that governs solely in the expansionary phases, without any political damage in the future. If, after winning the elections, the economy is hit by recession, the logic of fi scal cynicism dictates that the respective party steps down, so as not to pay the cost of the expansionary fi scal policy it promoted. The explanation I have put forward triggers immediately two questions: why the ruling parties that promote fi scal cynicism are not penalized by voters? And why the other parties do not adopt strategies to prevent the governing of one and the same party only in the expansionary phases of the business cycle? The answer to the fi rst question is that the lack of sanctions from voters owes to the infl uence of three factors. One such factor is none other but the aforementioned cognitive bias of the public at large, to whom the signifi cance of losses is greater when compared to that of gains of equal magnitude. Another factor is that, while almost all voters can understand during an economic upturn the immediate benefi cial eff ects of pro-cyclical fi scal policies, the majority cannot comprehend how the negative eff ects of these policies, which usually become manifest under a diff erent administration, during a downturn, are ascribable to the previous governance. Briefl y put, most vot- ers cannot understand that the negative economic eff ects in the contractionary phase are rooted in the pro-cyclical fi scal policy promoted during the expansionary phase. From a cognitive perspective, in the explanation put forward here, the administration is aware and voters are oblivious of the detrimental eff ects over the longer horizon of fi scal policy, meaning there is a cognitive asymmetry at work. Finally, the third factor is short memory. Most of the voters who understand would forget by the time of the following elections that the costs emerging in recession are partly rooted in the pro-cyclical fi scal policies pursued by governments during periods of expansion. The three factors listed here, i.e. cognitive bias, cognitive asymmetry and short memory, are mutually reinforcing. Regarding the second question, the answer is that the lack of a strategy of the oth- er parties to render it impossible for one and the same party to govern only during all business cycle upswings promoting pro-cyclical fi scal policies refl ects a coordina- tion failure. Essentially, the other parties (namely those that do not succeed in gov- erning during upturns) fail to come up with a coordinated refusal to govern in re- cessionary periods. In our view, the underlying reason for this failure relates to the fact that institutions in our democracy are weak. That is why the complex structure of incentives prompting the decision to assume power is deeply skewed in favor of decisions spanning the short-term horizon. For those who repeatedly fail to govern during upswings, the incentives to rule in recessions are so strong that those who have a chance to do it cannot adhere to a strategy implying abstention from govern- ing. This outcome is known to the party that governs only during all business cycle upswings by promoting pro-cyclical fi scal policies. The parties’ failure to coordinate, 27 in conjunction with the lack of sanctions from voters, generates – in democracies with weak institutions – a very serious consequence: a party that manages to govern from the very beginning in the expansionary phase of the business cycle and promotes pro-cyclical fi scal policies has the highest chances of governing during the next busi- ness cycle upturns as well, while it is optimal – for the reasons stated above – not to rule in recessions. The fi scal cynicism explanation implies various degrees of severity of the pro-cy- clical fi scal policy. To refl ect the maximum damage, the explanation is worded as fol- lows: in a democracy with weak institutions, the administrations in the expansionary phase of the business cycle use the pro-cyclical fi scal policy in order to cripple the ad- ministrations in the contractionary phase all the way to extinction. This severe form of pro-cyclicality might be the only one preferable for those who intend to govern solely during all periods of economic boom, since it leaves political adversaries with very litt le room for maneuver, while maximizing the public’s positive perception vis- à-vis the governance during the economic expansion. In Romania, evidence seems to support the idea that the pro-cyclical fi scal pol- icy pursued in all expansionary periods infl icted maximum damage on the parties that managed the eff ects during periods of recession and exit from recession. Each recession occurring after 1990 that had been preceded by an expansionary fi scal pol- icy ousted from the political stage either the party perceived as leading a governing coalition during the recession or the party that formed the government during the downturn. That is what happened to the Christian Democrat National Peasants’ Par- ty (PNȚCD) after the 1997-1999 slump, and to the Democrat-Liberal Party (PDL) after the 2009-2010 recession. Data in Figure 1 validate the hypothesis that, in Romania, the pro-cyclicality of fi scal policy emerges from the very start of governances in business cycle upturns and becomes more visible during the closing years of the respective tenures, as stated by our fi scal cynicism theory. At present, the ruling coalition in Romania is pursuing a strongly pro-cyclical pol- icy, as illustrated by Figure 1. Based on previous experience, it results that the party coming to power or leading a governing coalition after the 2020 parliamentary elec- tions, or even sooner, if a crisis breaks out, will have diffi cult problems to solve if the economy enters a recession. The respective administration will not only be unable to use the fi scal policy to take the economy out of recession and speed up the return of output to potential, but it will delay these processes because it will be coerced to cut budget spending and raise taxes. Such measures might trigger voters’ ultimate aver- sion, similarly to what happened after the parliamentary elections of 1996 and 2008. The founding elections of 1990 brought to power in Romania a party which, either directly or by supporting parties that did not have a parliamentary majority, gov- erned in virtually all expansionary phases of the business cycle, always promoting pro-cyclical fi scal policies, but almost never ruled when the economy was mired in recession. Readers should be reminded that it is not our aim to determine wheth- er governing solely during business cycle upswings and systematically promoting 28 pro-cyclical fi scal policies are part of a strategy or are the outcome of a cynical or benevolent intention. But, looking beyond intentions, the aforementioned fi ndings are the same as those that would come up if the respective party were to govern only in all expansionary phases and to implement pro-cyclical fi scal policies with the intention of gett ing rid of its foes, like in the fi scal cynicism hypothesis. If the trend persists, this particular party will become very big and will no longer have any real political opponents after 2024. A democracy where there is no political struggle will eventually be altered. Es- sentially, this is the direst political cost that strongly pro-cyclical fi scal policies could entail in Romania. 2.2. The ‘constrained responsibility’ hypothesis The other hypothesis I put forward in order to explain why the parties governing during economic upturns implement pro-cyclical fi scal policies is that pro-cyclical- ity advocates have the necessary knowledge to fathom the risks posed to infl ation and unemployment, as in the fi scal cynicism explanation, yet are benevolent. More precisely, the promoters of pro-cyclical fi scal policy are guided by the belief that the associated risks will not materialize before shifting to a correct countercyclical policy, later on, after they have reached the seemingly non-delusional objectives pursued, for instance, via tax cuts alongside signifi cant pay rises. This explanation assumes, similar to the fi scal cynicism theory, that detrimental eff ects (costs) may occur under the governance implementing the pro-cyclical fi scal policy in the expansionary phase of the business cycle. But, unlike the fi scal cynicism theory, it additionally assumes a ‘responsibility’ of the fi scal policy initiators towards the public and political adversaries. ‘Responsibility’ is manifested in the intention of a ‘timely’ reversal of the pro-cyclical nature of fi scal policy, i.e. prior to the outbreak of a crisis that would halt the upturn of the business cycle. The timely shift would allow fi scal policy to become countercyclical, meaning that it can be used in the contraction- ary period to foster economic growth, thereby facilitating the task of the incoming government, which can be one and the same with the government having promoted the pro-cyclical policy. This ‘responsibility’ does not show up out of the blue. On the one hand, it might be generated if there were evidence prompting the governing party to believe it could stay in power for two successive electoral cycles. On the other hand, it is triggered by the awareness of a high likelihood that the business cycle upswing might be interrupt- ed by the occurrence of a slump during that particular term in offi ce. We can now say that we are dealing with a responsibility ‘constrained’ by the prospects of recession. If we assume, without it being an assertion, that this knowledge and the belief in the ‘timely’ reversal of the fi scal policy nature – which I have already brought up – lie at the root of pro-cyclical fi scal policies in Romania, then shifting to the implementa- tion of the expansionary pro-cyclical fi scal policy as early as 2016-2017 would refl ect two implicit wagers. The fi rst is that monetary policy will somehow succeed in keep- 29 ing infl ation within the range targeted by the central bank, despite the relatively wide structural budget defi cit, without creating discontent for the public. The decision to tighten monetary policy is, however, not an easy one. On the one hand, if it did not raise the interest rate to tackle infl ation, monetary policy would contribute, as I am about to show, to the widening of the structural budget defi cit, while the leu would weaken in the anticipation that macroeconomic policies are unsustainable, which would be visible in infl ation. In this case, the infl ation wager would be lost and, from the perspective of those who bet on infl ation staying low, it would have costs in terms of future votes. Alternatively, infl ation might continue to stay low for a long period if, follow- ing the earlier structural changes or those recently generated in the aftermath of the 2008 crisis, the link between excess demand and infl ation has been greatly weakened. Thus, infl ation readings close to the lower bound of the variation band around the fl at infl ation target set by the central bank may coexist for a while with an infl ationary, even rising, output gap, without requiring a policy rate hike to narrow the output gap. That is why, alongside the widening of the private sector’s external imbalance, two highly costly consequences emerge from the relationship between monetary and fi scal policies. The fi rst is that, having no reasons given by prospects of higher infl ation to reduce the output gap to zero, monetary policy contributes to the strengthening of fi scal poli- cy pro-cyclicality. This eff ect occurs if the headline budget defi cit becomes an implicit target. Such a target appears when the adopted fi scal measures tend to raise the head- line defi cit above a certain limit, for several years in a row, but there are forces (such as the rules in the European Union) which impose sanctions that make the defi cit stay at the maximum admitt ed limit by rules. In this case, even if the headline budget defi cit remains constant for a number of years as a percent of GDP, the rise in the excess demand (GDP gap) leads to the widening of the structural budget defi cit from one year to another, as I have previously mentioned. In this way, by refraining from tackling excess demand, a pro-cyclical monetary policy in the expansionary phase of the business cycle would add to the severity of the fi scal adjustment in recession, contributing to higher unemployment. The second consequence, which I will dwell upon in the next section, is that a pro-cyclical fi scal policy may lead, with the entry into recession, to the loss of the monetary policy’s operational tool. On the other hand, tackling infl ation might require relatively large increases in the monetary policy rate in Romania. This would strain the balance sheets of the indebt- ed, likely to generate the emergence of the fi nancial frictions that will entail a slow- down in economic growth or, in the worst case, a recession. The second implicit wager I have referred to is that the expansion of the world economy will not come to a halt in the period when our fi scal policy is pro-cyclical and that, as it has happened in Romania so far, a recession will not be generated exclusively on domestic grounds. In other words, it is a bet that global growth will extend beyond the moment marking the deliberate return of domestic fi scal policy 30 to the countercyclical profi le. It is a bet on the approximate date of the outbreak of the following crisis (recession) in a large country, which may infl uence the business cycle globally, meaning that it is a bet against the economic cycle. This bet refl ects the belief that, within the same expansionary phase of the economic cycle, fi scal policy can fi rst be pro-cyclical, in order to att ain some goals, and it can then be turned into a countercyclical policy even before the upturn has ended, in virtue of the constrained responsibility. From another perspective, fi scal policy is pro-cyclical from the begin- ning in order to maximize the chances of not facing a crisis during the deliberate pe- riod of pro-cyclicality and to stand a higher chance of being turned from pro-cyclical into countercyclical in due course. This belief is, usually, illusory. The data I have presented for Romania in Figure 1 show that our fi scal policy has never been relent- lessly countercyclical. Briefl y put, I call this belief the illusion of the ‘timely reversal’ of fi scal policy from pro-cyclical to countercyclical. Since the illusion of the timely reversal cannot be done away with, the pro-cyclical fi scal policy pursued from the perspective of constrained responsibility would have the same negative economic eff ects for the following governance and for each and every one of us. However, it would not exert the same negative eff ects in the long run on democracy as fi scal cynicism, because presumably the party pursuing it does not wish to govern solely during expansionary phases of the business cycle. Nevertheless, this neutral result on democracy would be diff erent if we were to change the constrained responsibility hypothesis to allow one and the same party to govern, with the benevolent intention of ‘timely reversal’, only during all business cy- cle upturns. We thus derive the ‘altered constrained responsibility’ theory, which is a particular case of the constrained responsibility theory, where the respective party is benevolent not only to the public and political foes, as in the initial defi nition, but also to itself, since it avoids governing in recession. With these changes, the constrained responsibility theory yields the same negative results with regard to democracy as the fi scal cynicism theory. In order to be clearer on the similar negative results on democracy of both fi s- cal cynicism theory and altered constrained responsibility theory, I reiterate that the founding elections of 1990 in Romania made it so that, without governing during downturns, one and the same party governed directly or indirectly (by supporting a minority government) in virtually all business cycle upswings ever since 1990. As- suming this party has so far always pursued and were to pursue from now on as well pro-cyclical fi scal policies with the benevolence implied by constrained responsibil- ity, and were to continue to govern only during all expansionary phases of the busi- ness cycle, then it is highly likely the party will no longer have any real opposition starting 2024. It is the same result obtained in the logic of fi scal cynicism as well. Now we can clearly state that it does not matt er whether a party is cynical or be- nevolent when pursuing a pro-cyclical fi scal policy. What matt ers is if, by pursuing the pro-cyclical fi scal policy, the respective party has the strategy, without governing in any recession, to govern in virtually all business cycle upturns and manages to im- 31 plement it. Any political party or coalition meeting this condition will end up having no political adversaries that count. In this way, the said party or coalition may pro- foundly alter democracy. Having defi ned the two hypotheses – that of ‘fi scal cynicism’ and that of ‘con- strained responsibility’ –, it becomes clear that, in the cognitive approach adopted in this paper, if considering strictly a given governance, rejecting the ‘constrained re- sponsibility’ hypothesis automatically means accepting the ‘fi scal cynicism’ explana- tion, and vice versa. However, if we have in mind the case in which one and the same party governs for a relatively long period in all business cycle upswings, and does not govern during recessions, it cannot be ruled out that certain governances might have been guided by fi scal cynicism, while others by ‘altered constrained responsibility’. Irrespective of the correct explanation, the same party governing in all expansion- ary phases of the business cycle, without being in power during any recession, along with perpetually promoting pro-cyclical fi scal policies may lead to an alteration of democracy and, depending on the normal or abnormal character of the business cy- cle, to the loss of interest rate as a conventional monetary policy tool. 3. An assessment of risks of the pro-cyclical fi scal policy from the perspective of normal and abnormal phases of the economic cycle Even though fi scal policy conversion from pro-cyclical to countercyclical within the same expansionary phase of the business cycle is generally illusive, it cannot how- ever be ruled out. Starting from the assumption that the current political coalition in Romania is ruling guided by ‘constrained responsibility’, the aim of this section is to answer the question whether this time there are chances for the fi scal policy in Roma- nia to become countercyclical in the present expansionary phase of the economic cy- cle, thereby reducing the odds of losing the interest rate as a monetary policy tool. For this purpose, we shall analyze some traits of the global economic cycle and the condi- tions in which the next phase of the economic cycle could be ‘abnormal’ for Romania. 3.1. The inevitable recession We know for sure that a new crisis will come, but we never know when; in prin- ciple, a crisis is necessary as it shows us what is wrong in our economic model, i.e. in our business model. Provided we identify correctly what needs to be fi xed, we can lay the groundwork for a lasting expansion, in which imbalances build up slowly and economic growth is high. If we identify them incorrectly imbalances may build up fast and expansion will be brief. With a wrong identifi cation, even if the economy were to grow at a fast pace under the infl uence of incidental factors, the shorter du- ration of the upturn might make it so that the expansion-generated production gain trail behind the outcome of the correct identifi cation of the underlying causes and the adjustments needed. Admitt ing that the experience will recur, Romania’s economy will not enter re- cession before being forced by a crisis generated outside it. Such a crisis aff ecting the 32 domestic economy as well should erupt in a large country, with strong ties to the Eu- ropean Union, the main destination of Romanian exports. Such an economy is the US, which is the largest world economy and the leading trade partner of the European Union. Several parameters of this economy prompt us to believe it will not be long before a new crisis (recession) hits: a) the economic growth rate during the present expan- sion has been lower than in the previous two, which might be indicative of the fact that the US economy has not yet conducted all the adjustments required by the 2008 crisis, either because it has not properly identifi ed them or because it was unable to; b) non-fi nancial corporations embarked in 2011 on a new indebtedness cycle, with the annual pace of increase of debt exceeding 6 percent in 2015; c) the ratio of net debt to funds from operations has followed a steep upward path since 2015, and has exceeded 170 percent, above the 2009 reading; d) the average credit cycle spans 6-7 years, whereas the aforementioned expansion of credit to non-fi nancial corporations since 2011 shows that this cycle is drawing to an end of its average length; e) fi nally, although it is known that expansions do not die of old age, it is also worth mention- ing that the longest expansion in US history lasted for 10 years; the ongoing one has already been around for 8 years. For the time being, the negative signals coming from these data are alleviated by the fact that the credit cycle is barely at the beginning in the euro area and Japan, while in the US the government and households increase their debt at slow paces, of 5 percent and 3 percent per year respectively. Under these circumstances, it may be assumed that Romania’s fi scal policy might be caught on the wrong foot, meaning that it cannot be turned from pro-cyclical into countercyclical in due course, losing the implicit wager referred to above. Moreover, turning the fi scal policy from pro- to countercyclical cannot be done overnight, it takes time. This shows that economic expansion might be halted by a crisis before the ‘timely reversal’ of the fi scal policy. 3.2. The emergence of the ‘abnormal-recessionary’ phase and the loss of conventional monetary policy In order to evaluate the consequences of the pro-cyclical fi scal policy during the economic boom, we should correctly identify the phase of the economic cycle that the economy will be in. I am not referring here to the uncertainties associated with the econometric estimation of the excess demand, which is an unobservable variable. I am referring to the fact that the business cycle is more complex than it was generally believed prior to the 2008 crisis. In light of the experience, the economic cycle can no longer be judged as a succes- sion between the upturn and the downturn of the business cycle, in which monetary policy rates are signifi cantly positive. The crises – starting from that of 1929 and all the way to that of 2008 – have shown that the economic cycle has a ‘normal’ phase, in which policy rates are signifi cantly positive, and an ‘abnormal’ one, in which the said rates are zero or even negative. 33 In each of these phases, the periods in which the economy is in expansion alter- nate with periods in which the economy is in recession. Thus, the economy can be in one of the four possible phases: (i) normal-expansionary, (ii) normal-recessionary, (iii) abnormal-recessionary, and (iv) abnormal-expansionary. The macroeconomic policy mix is conditional on the phase of the economy, which makes it so that the key role is not always assumed by the same policy (Croitoru, 2016). An example of the four phases of the economic cycle for the US economy is shown in Figure 2. Having clarifi ed these phases, it can be seen that, once the economy enters a re- cession after a normal-expansionary phase, such as the one that started shyly in Ro- mania in 2011, and has already led to economy overheating in 2017, in which fi scal policy was pro-cyclical, the emerging eff ects depend on the phase that comes next. If the next phase is the normal-recessionary one, in the worst case fi scal policy might continue to be pro-cyclical during both the recession and much of the period when output would remain below potential, even if the economy grew at positive rates. But, very important for the reasoning here, monetary policy will have the necessary room to cut interest rates enough to help the economy exit the recession quickly. In other words, monetary policy will not lose its operational tool, namely the interest rate. -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 19 14 19 18 19 22 19 26 19 30 19 34 19 38 19 42 19 46 19 50 19 54 19 58 19 62 19 66 19 70 19 74 19 78 19 82 19 86 19 90 19 94 19 98 20 02 20 06 20 10 20 14 Interest rates on the secondary market for 3M T-bills (annual) Annual GDP growth rate Monetary policy rate (Fed rate) Consumer prices annual inflation % NP AP AP Source: author’s creation based on data available at FRED Figure 2: ‘Normal’ phases (NP) and ‘abnormal’ phases (AP) of the business cycle in the US Nonetheless, the normal-expansionary phase may be followed by the abnor- mal-recessionary one, as it happened with the 2008 crisis, when aggregate demand witnessed a massive drop in developed countries. In this case, the asymmetric con- 34 tribution of the pro-cyclical fi scal policy can lead to the loss of the monetary policy operational tool. On one hand, in the expansionary phase, the pro-cyclical fi scal pol- icy made a relatively small contribution to the increase in the interest rates, which created additional room for monetary policy. On the other hand, however, in the ab- normal-recessionary phase, fi scal policy may remain pro-cyclical, substantially con- tributing to the deepening of the recession and of the unemployment, calling for a major cut in the interest rate, as in the case of the normal-recessionary phase, or it may remain insuffi ciently countercyclical. Now, however, unlike the normal-reces- sionary phase, the interest rate cut to zero is no longer suffi cient to help the quick exit from the recession. Now, since it widened too much in the normal-expansionary phase, the fi scal defi cit either needs to be diminished or can no longer be suffi ciently increased to make up for the decline in private demand, with monetary policy having to try to boost aggregate demand via quantitative easing. The less indebted in foreign currency the economic agents are and the larger the stock of instruments available for such operations, the higher the success of such a policy in Romania. In the case depicted here, the monetary policy of quantitative easing is a conse- quence of fi scal policy pro-cyclicality. Or, otherwise stated, monetary policy loses its operational tool because fi scal policy was pro-cyclical in the expansionary phase of the business cycle. This scenario was confi rmed after 2008 mainly by the developed economies, but also by an emerging one: the Czech economy. In developed economies, the loss of interest rate as a monetary policy tool occurred because policy rates had been relatively subdued prior to entering the Great Recession of 2008, and fi scal defi cits could not be suffi ciently widened to make up for the fall in demand to levels at which the natural interest rate became too low. Essentially, the suffi cient widening of defi cits was impossible either because, in some cases, fi scal poli- cies had been pro-cyclical during the upswing or because of governments’ fear that an overly large increase in the defi cits might have not been fi nanced by markets, which would have also aff ected economic agents’ confi dence, already at very low levels. Looking at the Czech Republic, structural budget defi cits before the crisis had been relatively wide, so that the entry into recession for the second time after 2008 called for fairly large cuts in the structural budget defi cit, of 1.5 percentage points in 2011, 1.1 percentage points in 2012, and of 1.6 percentage points in 2013. In other words, the Czech Republic lost fi scal policy because it could not expand the budget defi cit in order to stimulate the economy. The Czech Republic’ central bank bought a total of 76 billion euros, which was tantamount to injecting korunas into the economy, i.e. it pursued a less common form of monetary policy based on quantitative easing. 3.3. Abnormal-recessionary phases will extend across countries and will become more frequent The discussion in the previous paragraph on pushing the economy by a pro-cycli- cal fi scal policy in the abnormal-recessionary phase and on losing the interest rate as a monetary policy tool was based on two hypotheses. The fi rst assumes a signifi cant 35 fl att ening of the Phillips curve at the low levels of the infl ation rate, including in Ro- mania. Were this fl att ening not to take place, infl ation would pick up at higher paces as output rises towards potential. The second hypothesis is that the natural rate of the interest might drop to very low levels. With these two assumptions, cutt ing the monetary policy rate to zero would not suffi ce to push the real monetary policy rate (which, in this case, is equal to zero minus infl ation) below the natural interest rate because infl ation would be too low. Arguments and evidences to support these hypotheses are available in Bernanke (2005, 2007 and 2015), Borio and Disyatat (2011, 2014), Borio (2017), Carney (2017) and Croitoru (2015). These papers show that global factors and technological changes lead to: (i) the decline in the interest rates and in the natural rate of interest, (ii) the change in fi nancial conditions, and (iii) the fl att ening of the Phillips curve. The materialization of these hypotheses is underway in Romania, and they might fully materialize in the next two to three years. During this time span, as I have argued above, the US economy may enter into recession, triggering a recession in Romania as well. With the aforementioned hypotheses materialized, and with a fi scal policy that seems programmed to stay pro-cyclical over the next two to three years, the upcom- ing phase of the business cycle is highly likely to be abnormal-recessionary, meaning that the interest rate can no longer play the role of a monetary policy tool. The low natural rates of interest and the fl att ening of the Phillips curve will prompt the interest rates and the infl ation rates in various countries, Romania included, to stick to relatively low levels. Against this background, if recession sets in, our econo- my will need a relatively large policy rate cut to resume growth. Since interest rates will probably be too close to zero, it is also likely that monetary policy might not be able to deliver the necessary reduction in the policy rate. After lowering the interest rate to zero, the real rate of the interest is equal to minus infl ation. If infl ation is low too, as we have assumed, then the real monetary policy rate is negative and relatively high (let us admit, for the sake of a comparison I am about to make, that it is, for ex- ample, minus one percent). From this point, the resumption of economic growth hinges on the level of the natural interest rate and on the fi scal policy. If the natural rate of the interest has dropped below the level to which the real monetary policy rate can be cut (minus two percent, for instance), then monetary policy (whose interest rate could only be low- ered, in our example, to minus one percent) can no longer contribute via conventional tools to stimulating demand. Now it takes expansionary fi scal policies to stimulate demand. This would shift the demand curve upwards enough to raise the natural rate of the interest to levels from which monetary policy would regain its strength with conventional means. The only thing is that, having been pro-cyclical during the economic upturn, fi scal policy is very likely to stay so in recession as well, unable to help with the resumption of economic growth and with the regaining of the mone- tary policy. This is the mechanism which may lead to the loss of monetary policy that I have referred to in the fi rst two sections of the article. 36 In this way, the pro-cyclical fi scal policy in the expansionary phase might push our economy into an abnormal-recessionary phase, which it can exit only by resorting to unconventional solutions. The loss of conventional monetary policy is the highest economic cost Romania could pay for promoting pro-cyclical fi scal policies. 4. Conclusion In this paper, I have shown that if, in a democracy with weak institutions, fi scal policymakers are aware of the potential negative eff ects associated with promoting a pro-cyclical fi scal policy, there are two explanations for accepting the pro-cyclicality of fi scal policy. When the ruling political coalition is not benevolent and succeeds in governing solely during virtually all normal-expansionary phases of the business cycle, ‘fi scal cynicism’ emerges: the pro-cyclical fi scal policy is promoted in the ex- pansionary phase for the temporary benefi ts it brings to the ruling coalition and for the diffi culties it poses, thereafter, to the political opposition. When the ruling politi- cal coalition is benevolent, a ‘constrained responsibility’ occurs: the pro-cyclical fi scal policy is promoted in the normal-expansionary phase with the intention that, after meeting some economic and political objectives, it should be turned into a countercy- clical fi scal policy well before the economy enters a recession. I have shown that the timely reversal of fi scal policy from pro- to countercyclical is illusive, so that negative eff ects emerge for those ruling in recession, as in the case of fi scal cynicism. Thus, the constrained responsibility turns into an ‘altered constrained responsibility’ if the benevolent political coalition succeeds in governing solely during virtually all nor- mal-expansionary phases of the business cycle. I have shown that, by practicing either fi scal cynicism or altered constrained re- sponsibility, a political party or coalition will end up dominating the political stage, without any real opposition. Both the theory of fi scal cynicism and that of the altered constrained responsibility explain well the disappearance of major parties from the Romanian political stage after the recessions of 1997-1999 and 2009-2010, and indicate that a single very large party may end up dominating the stage, thus altering the de- mocracy. It was not our aim to show which of the two theories bett er explains these outcomes, and which theory applied in various periods from 1990 onwards. The only real strategy that remains for avoiding the possible serious consequences of a pro-cyclical fi scal policy – erosion of democracy, loss of fi scal policy in recession- ary periods and, therefore, possibly the loss of the interest rate as a monetary policy tool – is to strengthen institutions, whose improved quality should stand as a guaran- tee that good fi scal rules are applied, preventing the emergence of pro-cyclical fi scal policies. References: 1. Bernanke, B.S., ‘The Global Saving Glut and the U.S. Current Account Defi cit’, Re- marks at the Sandridge Lecture, Virginia Association of Economists, Richmond, Vir- ginia, March 10, 2005, [Online] available at htt ps://www.federalreserve.gov/board docs/speeches/2005/200503102/, accessed on February 19, 2017. 37 2. Bernanke, B.S., ‘Global Imbalances: Recent Developments and Prospects’, Speech de- livered at the Bundesbank Lecture, Berlin, Germany, September 11, 2007, [Online] available at htt ps://www.federalreserve.gov/newsevents/speech/bernanke20070911a. htm, accessed on February 17, 2017. 3. Bernanke, B.S., ‘Why are Interest Rates so Low, Part 4: Term Premiums’, 2015, [Online] available at htt ps://www.brookings.edu/blog/ben-bernanke/2015/04/13/why-are-inter est-rates-so-low-part-4-term-premiums/, accessed on April 15, 2015. 4. Borio, C. and Disyatat, P., ‘Global Imbalances and the Financial Crisis: Link or No Link?’, Bank for International Sett lements, Working Paper No. 346, May 2011, [Online] available at htt ps://www.bis.org/publ/work346.pdf, accessed on February 17, 2017. 5. Borio, C. and Disyatat, P., ‘Low Interest Rates and Secular Stagnation: Is Debt a Miss- ing Link?’, June 25, 2014, [Online] available at htt ps://voxeu.org/article/low-inter est-rates-secular-stagnation-and-debt, accessed on June 31, 2014. 6. Borio, C., ‘Through the Looking Glass’, OMFIF City Lecture, September 22, 2017, [On- line] available at htt ps://www.bis.org/speeches/sp170922.pdf, accessed on September 25, 2017. 7. Carney, M., ‘[De]Globalisation and Infl ation’, IMF Michel Camdessus Central Bank- ing Lecture, September 18, 2017, [Online] available at htt ps://www.bis.org/review/ r170920a.htm, accessed on September 18, 2017. 8. Croitoru, L., ‘The Tendency Towards Secondarity in Managing Global Imbalances’, 2015, Scientifi c Annals of the ‘Alexandru Ioan Cuza’ University of Iaşi, Economic Sciences, vol. 62, no. 3, pp. 291-311. 9. Croitoru, L., ‘Ceva ce ar trebui reparat în gândirea noastră’ (Something That Should Be Repaired in Our Thinking), September 21, 2016, [Online] available at htt p://www. opiniibnr.ro/index.php/macroeconomie/161-ceva-ce-ar-trebui-reparat-in-gandirea-no astra?highlight=WyJjcm9pdG9ydSJd, accessed on September 21, 2016. 10. Croitoru, L., ‘Two Outstanding, Less Known Contributions that Wicksell Has Made for Us’, March 28, 2017, [Online] available at htt p://www.opiniibnr.ro/index.php/ english/197-two-outstanding-less-known-contributions-that-wicksell-has-made-for- us?highlight=WyJjcm9pdG9ydSJd, accessed on March 28, 2017. 11. Reinhart, C.M. and Rogoff , K., ‘This Time is Diff erent: A Panoramic View of Eight Centuries of Financial Crises’, NBER Working Paper No. 13882, March 2008, [Online] available at htt p://www.nber.org/papers/w13882, accessed on December 17, 2008. 12. Wicksell, K., ‘Interest and Prices: A Study of Causes Regulating the Value of Money’ [Original publication date: 1898], translated by R.F. Kahn, Macmillan, 1936.