Well thanks online and in the room for coming to this this paper is the result of a collaboration with Economist at the bank of Italy a lot of work involved because the data problems were quite serious historical data had to be spliced and matched with more recent data and anyway lot
Way but it fits with program that I’ve had for many years sorry there’s no sound coming through oh is that DAV yeah go can you hear us I don’t know what’s happening because we tested this before let’s try go oh she did okay that’s fine Dave
David I think that must be a problem at David and then if y can hear us is what we might to sorry yes good thank you anyway as saying it fits with a long-term research program to try to improve policy models so we know that the new kees and
DSU models um really failed terribly to capture the financial accelerator and finance really eony linkages but the semi-structural models that many central banks most central banks now have to replace their previous tsg models um are also quite defective um and so that this has applications both for capturing monit transmission and for setting
Macroed dential policy now this paper is about consumption uh and consumption is an important part an important locus of the household financial sector and interactions transmission so the standard policy models in that have consumption functions consumption function in them which they all do um they exclude varying credit conditions and
Typically they aggregate all assets into Financial or total net worth in other words all assets minus debt to capture the household portfolio either they exclude a housing allog together or they make very inappropriate assumptions about about housing moreover they emit many models emit permanent income or doesn’t handle it very
Well so one question is why pick on Italy to uh to study consumption I’ll explain it okay so there’s been there’s been a bit of a revolution in macro so the the micro the macro evidence has strongly rejected representative agent R life cycle model of households that operate without liquidity in CR
Constraints Virginia’s research on on agents in a context of Inc markets so it’s changed the theoretical foundations and there’s now a lot of evidence on the cash flow channel of transmission of monetary policy particular in economies where there are floating interest rates like in the UK you know in the UK
Much of the talk the negative effects of consumption of those people who are heavily in debt whose mortgages are resetting so most central banks have realized the need to to go for semi structural models more flexible models where the data can speak uh can speak better so unfortunately these models net
Worth is the only way that asset prices liquidity and credit shocks affect consumption um they ignore liquidity differences despite quite a large literature on money the easier money where the idea is that you weit the different types of money um by how different their interest rates are relative to zero because the
Higher the interest rate the less liquid that type of money typically is so the idea of of waiting um different types of money differently because of liquidity is is you know deep in the profession and yet T banks have not taken that lesson and more widely to look at
Portfolios these models trivialize the role of debt relative to housing and stock market wealth and they ignore shifting credit constraints so they miss what me and the Sufi call the credit driven household Demand channel and they haven’t got good stories for debt house prices and residental construction and
As a result they can’t explain the financial accelerator and also how different different countries respond to these kinds of shocks and they also miss major channels of monetary transmission so on housing and consumption the micro research has been shown that the main housing wealth effect is more of a
Collateral effect than a classical wealth effect so it’s countries with home equity withdrawal options where housing wealth really matters and it implies the housing wealth should be treated differently from other assets but apart from good neog re of doing that and when you think about the collateral role it should be
Time bearing because it depends on how easy it is to refinance your mortgage if you try to use equity withdrawal you know and and some sometimes it’s very easy to refinance other times it’s not so that time variation is an important part of the story now where home equity withdrawal is
Difficult and pred constraints are tight and say that the down payment constraint that that the regulators and the banks demand from households if they are large it’s actually possible for higher house prices other things being equal to actually reduce AG consumtion which is the opposite of the the conventional um
Us American us UK kind of assumption so I’ve been aring for a long time that we need a household um housing system where the consumption portfolio asset price housing investment um is modeled jointly and um we produced some papers on these lines for France for example and jine L is working on South
Africa um made a system of this exactly this kind so we need at the very least an equation for consumption for non-mortgage debt Mortgage Debt for liquid assets for house prices and for permanent income then added to that housing investment and in principle it would be nice to include the drivers of illiquid
Financial assets but that’s the stock market and that’s a complex story that doesn’t depend so much on on households so for some countries we found that it’s actually impossible to make to tell a coherent story about any of these things without taking credit conditions into account and that’s very
Much the case for the France and South Africa now why Italy well Italy is different Italy has very low levels of Mortgage Debt relative to national income and it’s had quite modest and gradual rises in debt over time that makes it possible to study the balance sheet effects on
Consumption of different kinds of assets and debt without the complication of having to take this complex system with credit conditions varying hugely into account so in other words what I’m saying is that for France the US the UK South Africa it’s impossible to tell a cerent story about consumption without
Taking credit conditions into account brly it turns out that you can so it turns out that credit conditions only have a short-term effect they don’t affect the long run solution for consum and as a result you can get a very nice get a very nice understanding of how the different liquidities of financial
Assets affect consumption now why is it that Italy is different in this way well for quite a few reasons The Innovation legal system in Italy where Banks trying to access natur Al um you know a mortgage that’s cized in the house might face six years of delays and
Actually getting a hold of the house because the legal system is so conversome and so slow the law is also very favorable to tenant and borrowers and not to landlords and creditors and of course the culture has adapted to this so of culture of pattern
Is to young adults to stay long in their parents home as a result when you get the data first in Italy apart from those inheriting from their grandparents um are older than most other European countries but just to show how different Italy is this is Italy this is the mortgage debt to income
Ratio um here’s Ireland this is the Netherlands Spain so Italy really is so show you some data on Italy so you get a s idea of the key facts so in Italy the the saving ratio or one minus the saving ratio the rati of consumption total disposible
Income so between 1990 and about 2000s there was this massive rise in the ratio consumption to Total disposable income um and therefore a fall in the saving ratio and then it’s fluctuated a bit Co of course course had massive short effects so one of the key aspects of
Model is to try to explain why this happened now in many countries what was going on here for a similar rise in the consumtion income Ra was credit liberalization and a credit gr not in Italy now we don’t use commercial disposable income we use scaled income which gives uh non-property income a
Higher weight which is consistent with economic economic theory there’s a lot of evidence that the M propensity to to spend out of Labor income is higher than it is out of property income but the overall pattern scaled income is is pay so so here’s some data on on on household deposits liquid
Deposits have behaved like this debt has gone up quite gradually in the UK it’s contrast um in this period here debt exceeded liquid assets by some margin and in Ireland by a huge margin so another countries the picture looks very very different um the the net asset position um net liquid asset position
Has remained very positive in in in thises so let me turn to the key elements of the portfolio that we have data on Italy is one of those countries that’s quite unusual in that households actually own a lot of government bonds and corporate bonds um in the UK direct ownership of
Bonds um would be on much smaller much smaller level so you can see that in this period 90 to 2000 there were was a big increase in um in bond in in the value of bonds and and volume of bonds and also mutual funds so they both Rose
Very sharply in this period and it turns out that that is quite important clue to to what’s going on this is housing wealth relative to to income um in Italy housing wealth actually when you look at the total portfolio housing weth dominates um the other elements pensions and unquoted shares pensions is
The green line unquoted shares the and this is what happened to housing affordability namely the ratio of house prices to to income so I look at this period here 90 to 2000 quite a different pattern from any of the Ang Saxon countries um so in in
The UK um in Canada in in the US over this period there was a huge rise in house prices relative to income in Italy that was actually actually a fall and yet this is the period where the consumption of income braia is shooting up so you it’s a totally different
Situation from that of the SE countries so in the end we aggregate assets into sort of convenient groups and we have net liquid assets which is liquid assets minus debt um semi liquid assets which is uh the the green line sorry and IL liquid assets which is the
Black line and then we’re going to try to tease outsi these different elements so just to backtrack how does one estimate how do one think about consumption well well I’m very much um in line with with David Henry on this Theory based models make simplifying assumptions we generally
Don’t know which of those assumptions are correct and how one deduces the right model based on those assumptions so for Applied work it’s a good idea to to start with an encompassing approach with a a model that’s more General than the the simple model that theory throws
Up but encompasses or nest rival models so each the special cases by imposing parameter restrictions can be thought of the special case of of a general model this is something David and his authors coauthors have have written about for for a long time so the simple textbook model um
Looks like this in in um in log approximate approximate form so it says the the log of the consumption to current income ratio can be written in in terms of the permanent income to current income and asset assets to income total assets where the theory pretends that that net worth is liquid
And notic constraints of that kind so what we’re going to do is to generalize that by splitting Network into its key elements each with a different coefficient which reflects a three notion which is that cash for example is more spendable than your pension you know cash and the in your
Hand is obviously more spendable than the pension so it’s absurd to aggregate the more internet worth we’re going to allow the coefficient on permanent income which is one here allow that to to be different from one and also apply discount rates to future income which are far higher than
Sexation because after all future is very uncertain and people discount the future more heavily if the wa is uncertain um we allow the potentially The Intercept to vary with credit conditions Al turns out important um and in principle we need to test for the time variation in the the
Has a collateral effect turn the the down payment constraint because with if there were financialization those constraints would shift and therefore would affect would affect the coefficients on those on those components in moment and then equilibrium correction of course so the encompassing model looks like this so I
Said um in in princip sorry in principle The Intercept term could be time varying with credit conditions here we allow um boring rate and the and the deposit rate both to be in there um this is Perman inome to current income potentially time varying because liberalization might make you more forward
Looking net liquid assets illiquid Financial assets housing wealth and then the offsetting effect of of housing affordability and then possibly demography effects and for Italy we distinguish in addition to illiquid financial assets semi liquids exra so just think about signs for people who own property an increase in housing wealth relative to
Income is positive for for their for their spending for people who aim to be homeowners or renters an increase in house prices rather to to low income makes how less affordable they have to save more you got these two offsetting effects from different groups in the population and
It’s important to take both into account um so of the short run Dynamics income uncertainty proxied by the changing the unemployment rate something I found in just about every country I’ve ever done this for and income volatility turns out to be relevant as well and well it’s important to test for
Chang in the normal interest rate in the UK the change in the normal interest rate would be an important part of the transmission mechanism um because of the cash flow constraint on on borrowers and itally because debt is sorry how do you I’ll come yeah that’s very important question
Um so let me just say something about about the theory and how this more General model actually explains what consistent with a lot of micro evidence so this very important paper by Crawley and KLA just come out in the a macro Journal um looks at micro dat for denar
And uh they show that the modle PS to consume um varies greatly across households and it depends very much on the asset position of the household so the NPC is highest for the asset po intermediate for the for those holding liquid not liquid assets and lowest for
The doubly assd Rich those who own both liquid and IL liquid Financial assets and the encompassing model um is very consistent with that because if you think about the marginal pensity to consume um it depends on on these different um wealth to income ratios and you can show
Because this coefficient is typically a lot bigger than that and bigger than that you can show that people who have a lot of liquidity will have a particularly low um mity to spend um anyway I want to but there but it’s very consistent with with the microeconomics and the
Heterogeneous agent View and the micro evidence of the agent view so let me turn to you about per so how do conventional models when they include permanent income which many models don’t um how do they handle permanent income well the Fus model which is the US Federal Reserve sem structure model
Has two versions of per income they’ve got a model consistent version of future income expectations and one based on not Anil a small v um to to drive to drive income and they find that actually the mod consistent version works very badly and the V works better but these approaches down
Structural breaks well now you know I bet you wasn’t the Italian household um that foresaw the global financial crisis and the subsequent European sovereign debt crisis um so how does one handle that we can’t just Lively Run the V that ignores the fact there a huge structural break
In the data uh you need to do something about that so what we do is try to replicate what in thetion might have done the competion obviously would have made the same same forecasting mistake before the crisis but after the crisis they would have adjusted a model to
Incorporate the structural break so we have a model that incorporates the strual braak but then to back out what the expectations would have been beforehand we have to remove the expectation of of the the break that came later that couldn’t have been anticipated so we have to correct
Um the the fitted model for the fact that these crisis were unfort so our P model has a horizon of 40 quarters 10 years it’s got a discount rate of 5% per quarter that is about 20% per year quite High discount of the future and then we assume a learning process that happens
Once the crisis occurs um people start learning about about the crisis and they adapt their expectation a future income they shift down their exploitation of future income but it’s not always an instantaneous process and we make some s assumptions about thats Yeah question there’s also a big change in political system and economic
Management in Mia Series in the 90s right and also the adoption of the Euro yeah do you treat these think the structural breaks with increasing CC yeah the the um the the early ’90s um well Italy was forced out of the ER in in ’92 along with with the UK and
There quite a major shock that happens around that time and the political crisis as well um we don’t we handled that essentially through through a dummy rather than by trying to build into the income expectation process what happened it was a temporary phenomenon and then growth kind of resumed um and if you
Wanted to set up a complicated and learning apparatus for relative short uh a short time break it’s just not worth it it’s better to use use a dummy to to handle that particular one um as far as 2000 is concerned I think you’re you’re right because obviously the accession to
The euro Zone had a huge effect on interest rates and on expectations generally on asset prices and that’s in the model so you know the fact that we we have these big gains in the stock market and a lower interest rate big gains in the value of bonds that that’s
That’s exactly what’s in the model in terms of the drivers of permanent income one of the most important components of that is the log ratio of the working age population to the total population our data are all in per capita terms when you think about it if the working population were to shrink
Rather to the total population um that means the available income for for the total per capita um it’s going to shrink other things being equal so that’s changes in the working age population rather to the total population it’s important part of the General Dynamics of of of permanent income and there time
Trends obviously post post GFC Trend shift there’s both a shift in the the slope and a shift in the level um that’s Incorporated International competitiveness the unemployment rate oil prices the stock market all of these things and then some short run effects are in this reduced form of
Forecasting model and this is what estimated permanent income um an actual permanent income defined in a 40-year horizon uh looks like so the black line is the log of actual permanent income I’m taking future income over the next 40 quarters into account um the fitted line is the one that incorporates the the
Dotted line red Lin one that incorporates the structural break that couldn’t have been anticipated and then this shows the the learning adjusted Red Line shows that just before the crisis over optimistic about can find about 3 and a half% so they overestimated future income growth future level of Perman income by about 3
And half% and then they gradually um learned and by 2012 quarter 2 the assumption is they fully learned that new circumstances applied so that’s how we do what about credit conditions um my co-authors managed to dig out some interesting information from the Italian Central Credit register which is a
Measure of of overall credit not for households but overall and it’s um the ratio of granted credit lines to used credit lines and then we can also look at the ratio of granting credit lines to a moving average of GDP and that suggests that yes um you know I mean comparing 90
With 2000 you can see there was a a fall in credit conditions and then a rise so not much change um but then later on in the 2000s um for the liberalization then of course a big contraction in credit conditions after the crisis so it seems like a plausible indicator but I would
Say it has no long run effects of model it’s only the short Dynamics so but it so comes out let me say a word about about the speed of adjustment because this specification and speed of adjustment um are very closely connected and not all economists understand that I know that JY
Does and they do of course but you know I was trying to explain this at a seminar at the back of France a couple of years ago and to the people who actually construct back of France model and they completely failed to understand the point anyway those speed of adjustment
In in an ECM context equili correction context are a typical circum of specification problems um and the Phillips curve is a great example so the Philips curve effectively takes the second difference of the log price level orward second difference of log price level so this High degree of differen makes
Non-stationary data or data that’s subject to to shifts look stationary um in fact inflation is it isn’t only structural shifts it’s also long-term um long-term trends relative price trends are effect inion and those are missing in the new can in Philips gr and so you know it’s garbage in and
Garbage out um another great example from the DS literature is this paper by leaper published in the a are no less which is the model excludes credit asset prices shifts in credit constraints other regime shifts leaves out oil prices and you know what you get but
What you get is a model that’s again pushed towards second differen so driven by the order equation for consumption which incorporates habits and the Habit parameter is close to one which means that the order equation basically takes the first difference of consumption with habits close to one it’s like the second
Difference of consumption so it has to do that to eliminate all the things that have been left out and so course monsters implies that households don’t get utility from consumption they get utility from the change in consumption if you believe that you’ll believe anything um okay so let me turn to uh
The Italian estimates because they illustrate um this General point that a misspecified model typically results in very low speeds of adjustment having a well specified model then increases the speed of adjustment so what’s going on in this First Column is that the driver of consumption as far as the portfolio is
Concerned is just net worth actually Logan net worth relative to income and the speed of adjustment is 005 so you know saying that that over period of year only about a quarter the end of four quarters only about a quarter less than important the adjustment is taking place if you
Include the change in credit conditions it just goes up a little bit slightly improved model um you have the ratio of net worth as which the log is even worse then if you improve the model by incorporating by splitting wealth into Financial wealth and housing and you include this housing this housing affordability
Effect well it’s quite a big Improvement you know it doubles the speed of adjustment to to 10% so that’s says that after a year you know about 45% of the adjustment is taking place but the biggest Improvement comes when you split liquid when you split Financial assets into liquid semi liquid and
Illiquid and then the speed of adjustment jumps to .28 and if you include one one further small Improvement which is to allow um the p as GFC Ro of per income to to shift a little bit um it rises to. 29 so with 0. 29 28 um well about 70% of the adjustment
Um one that um 80% of the adjustment takes place after four quarters so it’s radically different from from a number like that the mechanis model itself has a speed of adjustment assumption function as a speed of adjustment 007 um so our approach is saying something radically different about the
Speed to which consumption reacts to shocks and shifts in portfolio portfolios so what kind of estimates do we get well uh let’s take column five so that’s a model that’s got the three-way split of financial assets it’s got um housing wealth and it’s got affordability and you can see that the
Marginal propensity out of net liquid assets is about 0015 016 0 012 out of semiliquid assets 06 out of Il liquid assets we’re actually un shares have been removed entirely because they uh know it’s more significant um and then housing wealth Mar dep is around4 but it’s offset very
Substantially by the negative effect of of affability so the story about housing is complex and interesting the bank of vity model just drops housing all together um and in a way you can see why that is because these two things upset each other you know housing wealth and
House pric rad through income um get in the opposite direction and so it seems to net out to zero um so that’s what in their model but actually when you think about it truth is quite so in the paper we got a table that shows parameter stability estimat the sample over different
Periods um one comment about one of the controls in the paper uh is for the the pension reform so in 2012 I responded to your point including politics so at the end of the the sovereign debt classes Italy was in you know under severe pressure to to reform
Um and one of the reforms that was that was brought in a very serious reform first really serious reform of the pend system was in 2012 and this this amount of evidence that examined the impact on the labor market on uh on other aspects of of behavior um
Suggesting that it had quite serious consequences now the table one um suggests that that uh it resulted in a long-term 4.9% fall in the consumption of income ratio which is a bit I think a bit impossibly high so one of the things that we we looked at was other other things going
On there long-term trends that been left out and one of the possibilities is demography and you can get a slight demographic effect it’s not significant but it does have an impact on on the estimate of this pension reform and it’s possible that that the r of price of durables um
Might also be part of the story so the reasoning here is that is that over time um consumption has has moved quite a long way away from real objects um into you know into the internet um into accessing um films and music and so on very cheaply um in a way
That wasn’t possible before so one can argue that that as technology which is quite well proxied by the rising price of durable durable goods have fallen in relative price to non durables fa systematically over time that technological change has shifted consumption away from real things that
You need to save up for to things that you don’t have to save up for and therefore might have might have had an impact on the on the consumption to income ratio it’s a possibility anyway you know it’s it’s a marginal effect but it does result in more sensible um
Estimates of the Pion effect um amazing enough the short run controls the three impulse dummies for um income growth LED income volatility change the unemployment rate change in public consumption short-term effects when you leave the M out there has almost no impact on the long run solu solution um which is pretty important
And impressive because when referees of paper like this will say oh you f the data you put in so many so many special things that you can’t really believe in lot need the more now changed um estimates a robust to instrume in current income and he might have thought well an
Alternative asset grouping would be worth looking at let’s suppose we aggregate semi liquid defined by bills bonds quot shares mutual funds let’s suppose IL liquid assets are unquoted shares and pensions well with that split a slightly worse fit slow speed of adjustment but the story about liquid
Assets debt and housing was complet down change so it’s robust as far as that goes so let me go back to hogen because I’m Marg at the beginning that the story of hetrogeneous Agents is is the big change that’s happened in mcro that’s Chang this transform the way we think about about Monet
Transmission um so argu the model is consistent with this mod modern heniz in complete Market view um Al interesting to note that when you the kind of data we’re using has lot distributional information in it instead of that being a negative something that makes aggregation more difficult and makes the representative agent more
Impossible well actually think of it as exactly the opposite um by disaggregating assets in this way and also incorporating effective Unemployment uh you’re capturing some of the heterogenity that’s that’s that’s in the system um so on the incident of unemployment the inue of different Assets in debt is
Very heterogeneous and you can see that so even this is data from the from the the the wealth income and wealth survey so even for household deposits um look at the income quintile 50 more than 50% of household depos liquid deposits are owned by the by the wealthiest quintar but the other
Assets it’s it’s even higher um you know equities manage investment schemes and so on even higher so there’s a huge amount of inequality and heterogeneity in the underlines the Port Port yes I would to this that also in Italy itan service all these these figures are under reported
In terms of financial we so I mean probably you anticipate about next point because yeah um part of the interest of comparing 2016 with 2020 is to see whether postco this is postco well so late 2020 the is December 2020 so at the end of the first year of Co
Obviously there was a big a big boost in what’s called excess saving over this period and it’s interesting to think well who held the the additional saving was it the wealthiest or was it kind of Fairly equally distributed across the population and you know actually when
You look at it it’s quite ambiguous it’s not obvious that it’s all the highest income households that had had the biggest increase in in their wealth Holdings but respond to your point even for deposits the 2016 survey um only captures 35 the total deposits that come from the survey only
Captures 35% of what the agre balance sheet is showing uh for deposits so you know My worry is that yes this distribution inflation is very nice but it may be actually completely misleading about what’s been happening to the underlying distribution because the Under reporting
So okay almost uh at the end let me say a few words about how one thinks of these Mar propensities so they incorporate the intrinsic aspect of liquidity transactions cost and price uncertainty but they also incorporate behavioral differences distinct owner types lower NPCs for the more affl and
So when if the average NPC were deposits at you know around 15% um it’s almost certain to be much higher than that for those for gaset poor and lower than that for the asset Rich so in other words what we’re estimating is an agate effect and the distribution there’s a lot of
Distributional stuff underneath it that we don’t know about so this this literature on on uh on excess saving so this paper by by thisi shows that in the US um the the surge in Saving during the pandemic mostly in liquid assets in the Euro area much of it went to liquid or less
Liquids and they’ve got some estimates of implications I think for ital the implications wouldn’t be very good because they ignore the fact that actually what they the liquid acids also includes semi liquid acids which had very different behavioral consequences from from the most IL liquid assets soting illiquid assets into one Lum
Which isn’t very good so finally what do I conclude one of the amazing things is that you know the completely no braider inside everybody I’ve ever spoken to thinks it’s obvious that liquid assets are more spendable than pension wealth I don’t I have not found a single Central Bank model that incorporates
That simple idea not one why well it’s the micr foundation ideology of the rational expectations representative household um with the single liquid asset in a world without asymmetric information credit or liquidity because that gives you attractive optimization problem you know something you can solve with a solved
Out Solution that’s simple uh so that’s that’s what the models work with um and the other stupidity is even the intertemporal Neal model says that housing wealth is different from Financial wealth because housing is a consumption good whereas Financial assets are not directly consumed and so you should never agregate housing wealth
With financial wealth and yet most of the models actually lump housing wealth in in N worth so our finding is that summarizing househ of wealth wealth portfolios by net worth is a disaster than modeling consumption um the bank of Italy consumption function uh speed adjustment 07 um compared with ours so a very
Distorted picture the speed also the incidence of Monet transmission our model would say actually there are important long-term or medium-term Consequences that feed through from the portfolio position of households um pass prices are important but you’ve got to take into account the fact that they have the this
By a different effect for for borrowers um or potential um potential um owners of of housing people are saving for a deposit or renters than for owners um and debt is actually much more negative the effect of debt onction other things be equal more negative than the fincial models would would tell you
And so when you think about manage transmission you’ve got to think about these medium and long-term consequences um as well through the portfolio position as well as the direct effect of interest rates on consumption so that’s that’s the end of the story but you know handing housing correctly um
Splitting Financial assets is really key to getting good models and models that help you understand what’s going on help you um understand man transmission and help you with with financial stability issues because you if there is a housing story to the financial stability problem that the country faces or the banks face
Then it’s important to have a coherent story of of how that works so let me stop there questions comments yeah thank you very much that was really interesting um just wondering what do you think the policy implications would be um given that Italy has a very different structure uh to other European
Countries in terms of um control of the monetary policy and what other the policy options you think it should undertake given more evidence yeah that’s of course back in 1998 yeah um I wrot the paper with some colleagues about um about becoming monetary Union and worrying about the asym the asymmetries
Between countries um but the institutional asymmetries um that were there and the labor market asymmetries and the different history inflation histories and so on um and worried monetary Union between countries with such very different um structures could could result in very serious problems and um I remember about that time um a whole
Bunch of Treasury UK treasury Economist descended on nfield um and they wanted to discuss the the five IC tests because at the time um that the government the government at the time was considering whether to join the Euro or not and they proposed five economic tests that had been satisfied
Um entry conditions and of course I argu that for the UK it would be complete disaster unless we radically transformed um our financial system and a bunch of other things um because I said what would happen to Ireland subsequently I was afraid that that would happen in the
UK we would have an explosion of of house prices and debt um going out of control and then the financial crisis that follow um now Italy um didn’t have the credit driven crises that Ireland and Spain had but what it had was a a debt driven government debt driven
Financial crisis um the sovereign debt crisis and it still saddled with with huge amounts of of government debt um which is you know one of the one of the danger points that EUR places in this very uncertain World we’re in now um so you know having a cgy policy um for Italy
Is is a bit can be a bit of a problem I’m not sure it’s a problem at the moment um because the you know some big lessons have be learned as a result of the sovereign debt crisis and the European Central Bank is now standing by to potentially provide liquidity support
And to buy more Italian debt if if the stresses if the differences the spreads between German and Italian um s bonds rise too much EB would step by to to deal with that so you it’s being handled um but but obviously the the impact of inflation and interest rate changes it’s
Going to be rather different on Italy than than in other countries and probably the sector you know in ity probably Bears more heavily on on firms other than households households relatively low debt um anyway so that’s the kind of things you need to think about when when thinking about the situation
In the the current policy