I have made a model of an isle with few citisens and few institutions and limited natural resources.
There are maybe mistakes in file and I cannot conclude much from it for now but I will put it here. To view file you have to press button "zoom to fit" first because it is some problem in current version of Minsky ( 2.8).
I will put my latest file here. I concluded so far now that what some people say that government don't have to tax to spend is not true at least under Maastricht rules. If government has not enough taxes it would collect debt and net wealth would be redistributed to financial sector (owners of bonds) through interest rate on bonds. If Central bank buys government bonds from bank it is somehowe contradictoy to reserves requirement ratio. And if bank spend that additional reserves it could maybe be contradictory to its capital requirement (basel III and so on -I didn' read much about that). Maybe it could be masked with long term assets and depreciation rate but I don't have long term asset in model so far now and my system has limited natural resources.
If government has enough tax system goes to something which looks like equilibrium with horisontal lines. In that settings Citisen1 usually has positive net wealth and it is because I added material production to his asset and net wealth. But with that setting he always has more debt than money. Sum of net wealth is in my model equal to production stock because I raised agricultural production to "fair value"according to IAS41 (international accounting standard for agriculture). I have read somwhere that it is rarely done in reality and agricultural production is usually recorded based on historical costs.
With parameter TaxWealthorTaxSpendorTaxProfit I can switch between 3 type of taxes. With parameter InterestOnInterest I can switch between calculating interest on interest (continous compunding) or not calculating interest on interest. At least I believe that I manage to set that. With parameter DonationSelectcase I can add some free money from CB to government but limited with goverment maximum money parameter which is intialy 15. Goverment minimum money is intialy 10 and is important parameter. I tested adding new money few times and sum of net wealth was the same as before because it lowers CB equity and raises Goverment equity.
I maybe have mistakes in model and it is to some degree out of my control. I had difficulties in understanding time constants but used them because of manual which I have read.
Sum of net wealth stayed the same with new inputed money and equal to production stock because I inputed flow variable DonationCBtoGov in Central bank table differently from other types of money. I input it as negative on equity column for simplicity. If I inputed it as liability as other types of money - goverment money, reserves and cash - sum of net wealth would probably raise. And that makes me to think: are Central bank liabilities, liabilities at all, in one state system.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Interestingly, if I move flow variable DonationCBtoGov in Central bank table to liabilitie instead of equity where it is now it would be negative, unlike Cash1 variable which is positive.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
With tax on spend lowered from 0.2 to 0.05 it looks like this. Black line in chart Money and debt suddenly starts to grow at year 125 which I noticed many times. It happens because bonds are raising also.
And in my model loan is dependant on production because bank usually want some colateral when loans. I think that unsecured loans are more rare. Citisen1 (agricultural) has to borrow because it is not easy to pay tax with potato. I set it like that because it first came to my mind and thought that I will later check more if that is right or not but it stayed like that.
Citisen1 could be later agricultural sector and Citisen2 (which I didn't set) could later be some other sector - in example construction sector with long term assets or somehing else.
Last edit: Kresimir 2018-12-13
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I simulated in my file something which looks like Wynne Godley balance sheet matrix and transaction flow matrix. I think it is not without importance to understand that tables better, to see if there could be some simplification from them. I mean, if variables could be entered directly in them, or in some similar, new invented table - and without use of other tables, but with using some more rough equtations, it could maybe simplify model. I will maybe later delete them, because now they complicate my file a lot.
I called "transaction flow matrix" an "income and expenditure matrix", because word transaction seems to me misleading. In my opinion it should be viewed as something which happens on equity side, especialy when signs of variables from sources of flow variable changes are viewed - derivatives of other assets and liabilities stocks which I entered with ending sign ^d . I also added some things in that table which are not money transaction.
I started my file with agricultural product raising to "fair value" which is rigidly entered as 100. When I find time, I plan to make that "fair value" dependant on amount of money in system and maybe even amount of debt also, try to determine rate of inflation, input feedback of it to interest rates and add growth to material production which I plan to move to material production table. I will also check equtation for Lend variable.
I believe that, even with this simplified view, it can be felt what is happening in connected accounting of government, central bank, agricultural sector and financial sector..
I also notice, so far now, that when I add "new money" in the system that money usually ends on central bank liabilitie side - raising bank reserves above requirement ratio..
My latest file. I simplified tax on net wealth to only one tax parameter instead of four parameters and add posibilitie to calculate three taxes simultanously with tax on net wealth because of Picketty. I also added that tax on profit is calculated only to positive profits but there is not much change. I was unsure about that. Tax on net wealth change something but I run it only few times.
I have a feeling that there is not some ideal solution to monetary problems and that it is more about finding least worst solution and that it is much about tax system and regulation of financial sector. I cannot prove it now. But at least I tried.
My model don't follow strictly accounting standards and that is why objects which would not have accounting in reality have it in model. Table balance sheet matrix, income and expenditure matrix and nature matrix with depended variables left to it could be easily deleted, I believe, because their only function is similar to chart function.
I suppose that if I want to come from this to "minsky moment", in which I believe it is true, that I should put that production is dependant on lending, instead of opposite setting in model, and that that might be the case if there are unused recources in system and when that resources could be activated that way. It would maybe be more close to reality.
But my file is meant only to understand accounting better for now.
I also meant to inspect tesaurisation of gold by Central Bank. Is it usefull or human labour is captured that way not neccesarrily. And to inspect tesaurisation of foreign currency and to examine where it is held.
I don't have enough time for that now.
Since I didn't add so far now that value of production is dependand on amount of money or demand in system, I view file as negative photo with agricultural production only true money but money with which tax can hardly be payed.
As I know, according to IAS 41 someone who grows cattle can show biological production raised to "fair value" - market value. And it can show it as short term asset.
It is not the case with production of plants where only trees can be raised to "fair value" and showed as long term assets but fruits cannot be raised to "fair value".
I am not very familiar with accounting of other types of production but as I can recall they are recordered at historical costs including labour.
I added this only so that I can have some kind of production in accounting table.
It is sometimes thought that some combination of traditional accounting, where assets are recorded at historical cost, and some modern accounting where assets are valued with "fair value" is the best. And I meant to use it like that in file. Too much use of modern accounting and "fair values" is sometimes connected with malversations.
It could be maybe interesting to simulate production of oil similar like this. I only read so far now that it is recorded differently in North america and Europe.
Last edit: Kresimir 2019-01-17
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I tried to avoid negative variables with Heavyside function and sign < connected to stock variable and creating new variable with value 0 or 1 and then multiplying some other flow variable with that. In my opinion time constants are maybe not enough to avoid negative values because all values in tables are doubled. They could maybe be enough to avoid negative values if other side of change is always equity and equity allowed to go negative. But when there are, in example, concentric change (assets+ and assets -) or periferic change (liablitie + and liabilitie -) or centrifugal and centripetal change without change in equity (assets+ and liabilitie +) and (assets- and liabilitie-) they are maybe not enough to avoid negative values.
Last edit: Kresimir 2019-10-23
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I repaired few things in file - add two parameters for goverment money maximum so that subvention to citisen and donation CB to goverment can work at the same time. I also added one parameter for interesting rates that dissapeared from file.
Now the model is setted to add some money from Central bank to Goverment and with ability to give that money as subvention to citisen. What amount would it be is questionable. So I setted something but it is not what I meant at first.
It raises reserves ratio as in case when central bank buys goverment bonds.
And for that I lower tax on spending from 0.2 to 0.1 so that goverment can collect debt and raise ratio of goverment debt buyed by CB. Reserves are again raising with govement defaulted several times.
Again, my file maybe contains mistakes.
But in second case financial sector benefits.
That raising of reserves make me to think about Irving Fisher.
I am not sure how to measure inflation. It would be function of a money (which is dependand on loans) in the system and tau for spending. That would be demand.
If I add changable prices that would also influence on interesting rates. I mean, if citisen1 raise price of its product that inflation would raise interesting rates. And would even raise amount of loans because I set that loans are dependend on value of production, which I rigidly set as 100 per year. I set that lending is dependant on production because I started model from limited natural resources. In reality, it could easily be oposite.
I will leave it like this for some time.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I will put my latest file here. These time I am adding d/dt of interests calculated in financial sector to the money supply multiplied by some parameter. In these case that paramater is 10. I am adding it through parameter Donation_select case (left from Central bank table).
It was one of my major idea before I used the software so I wanted to simulate that. But I didn't expected that about parameter. Also, fraction reserve ratio rose from 0,1 to 0,2 which I didn't expected.
With that settings I get small inflation and Citisen 1 (agricultural) always has more money than debt, which I desired. I had to replace operator < with max function in some cases to be able to use d/dt operator.
I find that some conditions are stronger and some are weaker and tried to set conditions to be strongest about government money stock (GovermentM) and banks fractional reserve ratio (Reserves_requirement ratio). My opinion is that assumption that transactions are done imidiately is never true. All subjects will have some stock of money and that money has to be borrowed. Economical subjects to have zero stock of money is some boundary case leading to invalid conclusions.
Once I read Wynne Godley book to the end I will delete uncessary tables - balance sheet matrix and Income and expenditure matrix. They are my help tool for understanding that book.
My notes are sometimes not accurate but I left them because I see accountants as potential users of software.
Wynne Godley stressed importance of understanding of inflationary accounting but I didn't came to this point yet. Not even to pricing.
I guess that that parameter of value of 10 should have unit of year. First thing i thought about is 1/r i.e. 1/(interest rate). With high inflation then money added to supply would be lower and with lower inflation it would be higher. But I would further lower money added to supply with some other dimenesionless parameter (or combination of two parameter - in example velocity and something else) because I don't want hyperinflation in model.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I also used abs operator on one place without much need for it. It would make model dimensionaly not valid. It can be omitted or multiplied with number 1 of oposite unit. It want change anything.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Inflationary accounting I could maybe add with additional row in each table with flow variables correcting the value of stocks (reevaluating) but I want do that before I get rid of balance sheet matrix and income and expenditure matrix.
I have not previous experience in modeling so I setted conditions "by eye". When value of some stock is too low or too high, subject or "sector" will do some action -in example buy bonds" in amount related to "values" of other variables from its financial statement because it will be its first information. I counted that if I connect all other variables, that will trigger other conditions. It is maybe not the best way to run model but only I could do for now. That is the reason why equtation tab is hardly readable. Parameters outside model are, in example, production and interest rates an tau variables. I will probably try to make interest rates and also tau variables changeble with inflation to see if it is true that goverment has some inflationary gain on its debt with inflation.
Lending and repaying variables I imagine as more small loans and not one loan which is payed with annuity.
As I have already written, my model is meant to better understand accounting.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Another limitation is that citisen1 can take more loans than he need. So maybe stocks of money would have to be related to stock of production or maybe to "unused resources" or potential production. I will see if I will try to improve that. But based on this file I now understand CB financial statement, aggregated banks financial statements, sectoral financial statements and some other things better than before.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
And that thing about used and unused resources and production and potential production I see as something which could explain credit crises at least roughly. Since loan creates money but payments with interests comes later.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Because "techology" or using of a "resourses" don't improve continuously but improves with "steps" and financial system has other nature.
Fractional reserve system and bank liquidity ratio and capital ratio should also be more examined and interest rates and risks banks take compared to other subjects.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have made a model of an isle with few citisens and few institutions and limited natural resources.
There are maybe mistakes in file and I cannot conclude much from it for now but I will put it here. To view file you have to press button "zoom to fit" first because it is some problem in current version of Minsky ( 2.8).
Last edit: Kresimir 2018-11-25
I will put my latest file here. I concluded so far now that what some people say that government don't have to tax to spend is not true at least under Maastricht rules. If government has not enough taxes it would collect debt and net wealth would be redistributed to financial sector (owners of bonds) through interest rate on bonds. If Central bank buys government bonds from bank it is somehowe contradictoy to reserves requirement ratio. And if bank spend that additional reserves it could maybe be contradictory to its capital requirement (basel III and so on -I didn' read much about that). Maybe it could be masked with long term assets and depreciation rate but I don't have long term asset in model so far now and my system has limited natural resources.
If government has enough tax system goes to something which looks like equilibrium with horisontal lines. In that settings Citisen1 usually has positive net wealth and it is because I added material production to his asset and net wealth. But with that setting he always has more debt than money. Sum of net wealth is in my model equal to production stock because I raised agricultural production to "fair value"according to IAS41 (international accounting standard for agriculture). I have read somwhere that it is rarely done in reality and agricultural production is usually recorded based on historical costs.
With parameter TaxWealthorTaxSpendorTaxProfit I can switch between 3 type of taxes. With parameter InterestOnInterest I can switch between calculating interest on interest (continous compunding) or not calculating interest on interest. At least I believe that I manage to set that. With parameter DonationSelectcase I can add some free money from CB to government but limited with goverment maximum money parameter which is intialy 15. Goverment minimum money is intialy 10 and is important parameter. I tested adding new money few times and sum of net wealth was the same as before because it lowers CB equity and raises Goverment equity.
I maybe have mistakes in model and it is to some degree out of my control. I had difficulties in understanding time constants but used them because of manual which I have read.
Last edit: Kresimir 2018-12-13
Sum of net wealth stayed the same with new inputed money and equal to production stock because I inputed flow variable DonationCBtoGov in Central bank table differently from other types of money. I input it as negative on equity column for simplicity. If I inputed it as liability as other types of money - goverment money, reserves and cash - sum of net wealth would probably raise. And that makes me to think: are Central bank liabilities, liabilities at all, in one state system.
Interestingly, if I move flow variable DonationCBtoGov in Central bank table to liabilitie instead of equity where it is now it would be negative, unlike Cash1 variable which is positive.
With tax on spend lowered from 0.2 to 0.05 it looks like this. Black line in chart Money and debt suddenly starts to grow at year 125 which I noticed many times. It happens because bonds are raising also.
Last edit: Kresimir 2018-12-13
And in my model loan is dependant on production because bank usually want some colateral when loans. I think that unsecured loans are more rare. Citisen1 (agricultural) has to borrow because it is not easy to pay tax with potato. I set it like that because it first came to my mind and thought that I will later check more if that is right or not but it stayed like that.
Citisen1 could be later agricultural sector and Citisen2 (which I didn't set) could later be some other sector - in example construction sector with long term assets or somehing else.
Last edit: Kresimir 2018-12-13
I simulated in my file something which looks like Wynne Godley balance sheet matrix and transaction flow matrix. I think it is not without importance to understand that tables better, to see if there could be some simplification from them. I mean, if variables could be entered directly in them, or in some similar, new invented table - and without use of other tables, but with using some more rough equtations, it could maybe simplify model. I will maybe later delete them, because now they complicate my file a lot.
I called "transaction flow matrix" an "income and expenditure matrix", because word transaction seems to me misleading. In my opinion it should be viewed as something which happens on equity side, especialy when signs of variables from sources of flow variable changes are viewed - derivatives of other assets and liabilities stocks which I entered with ending sign ^d . I also added some things in that table which are not money transaction.
I started my file with agricultural product raising to "fair value" which is rigidly entered as 100. When I find time, I plan to make that "fair value" dependant on amount of money in system and maybe even amount of debt also, try to determine rate of inflation, input feedback of it to interest rates and add growth to material production which I plan to move to material production table. I will also check equtation for Lend variable.
I believe that, even with this simplified view, it can be felt what is happening in connected accounting of government, central bank, agricultural sector and financial sector..
I also notice, so far now, that when I add "new money" in the system that money usually ends on central bank liabilitie side - raising bank reserves above requirement ratio..
I also added navigation bookmarks.
My latest file. I simplified tax on net wealth to only one tax parameter instead of four parameters and add posibilitie to calculate three taxes simultanously with tax on net wealth because of Picketty. I also added that tax on profit is calculated only to positive profits but there is not much change. I was unsure about that. Tax on net wealth change something but I run it only few times.
I have a feeling that there is not some ideal solution to monetary problems and that it is more about finding least worst solution and that it is much about tax system and regulation of financial sector. I cannot prove it now. But at least I tried.
My model don't follow strictly accounting standards and that is why objects which would not have accounting in reality have it in model. Table balance sheet matrix, income and expenditure matrix and nature matrix with depended variables left to it could be easily deleted, I believe, because their only function is similar to chart function.
I suppose that if I want to come from this to "minsky moment", in which I believe it is true, that I should put that production is dependant on lending, instead of opposite setting in model, and that that might be the case if there are unused recources in system and when that resources could be activated that way. It would maybe be more close to reality.
But my file is meant only to understand accounting better for now.
I also meant to inspect tesaurisation of gold by Central Bank. Is it usefull or human labour is captured that way not neccesarrily. And to inspect tesaurisation of foreign currency and to examine where it is held.
I don't have enough time for that now.
Last edit: Kresimir 2019-01-17
With tesaurisation, which is maybe french word, I meant hoarding.
Since I didn't add so far now that value of production is dependand on amount of money or demand in system, I view file as negative photo with agricultural production only true money but money with which tax can hardly be payed.
As I know, according to IAS 41 someone who grows cattle can show biological production raised to "fair value" - market value. And it can show it as short term asset.
It is not the case with production of plants where only trees can be raised to "fair value" and showed as long term assets but fruits cannot be raised to "fair value".
I am not very familiar with accounting of other types of production but as I can recall they are recordered at historical costs including labour.
I added this only so that I can have some kind of production in accounting table.
It is sometimes thought that some combination of traditional accounting, where assets are recorded at historical cost, and some modern accounting where assets are valued with "fair value" is the best. And I meant to use it like that in file. Too much use of modern accounting and "fair values" is sometimes connected with malversations.
It could be maybe interesting to simulate production of oil similar like this. I only read so far now that it is recorded differently in North america and Europe.
Last edit: Kresimir 2019-01-17
I tried to avoid negative variables with Heavyside function and sign < connected to stock variable and creating new variable with value 0 or 1 and then multiplying some other flow variable with that. In my opinion time constants are maybe not enough to avoid negative values because all values in tables are doubled. They could maybe be enough to avoid negative values if other side of change is always equity and equity allowed to go negative. But when there are, in example, concentric change (assets+ and assets -) or periferic change (liablitie + and liabilitie -) or centrifugal and centripetal change without change in equity (assets+ and liabilitie +) and (assets- and liabilitie-) they are maybe not enough to avoid negative values.
Last edit: Kresimir 2019-10-23
I repaired few things in file - add two parameters for goverment money maximum so that subvention to citisen and donation CB to goverment can work at the same time. I also added one parameter for interesting rates that dissapeared from file.
Now the model is setted to add some money from Central bank to Goverment and with ability to give that money as subvention to citisen. What amount would it be is questionable. So I setted something but it is not what I meant at first.
It raises reserves ratio as in case when central bank buys goverment bonds.
Last edit: Kresimir 2019-02-24
And for that I lower tax on spending from 0.2 to 0.1 so that goverment can collect debt and raise ratio of goverment debt buyed by CB. Reserves are again raising with govement defaulted several times.
Again, my file maybe contains mistakes.
But in second case financial sector benefits.
That raising of reserves make me to think about Irving Fisher.
I am not sure how to measure inflation. It would be function of a money (which is dependand on loans) in the system and tau for spending. That would be demand.
If I add changable prices that would also influence on interesting rates. I mean, if citisen1 raise price of its product that inflation would raise interesting rates. And would even raise amount of loans because I set that loans are dependend on value of production, which I rigidly set as 100 per year. I set that lending is dependant on production because I started model from limited natural resources. In reality, it could easily be oposite.
I will leave it like this for some time.
I will put my latest file here. These time I am adding d/dt of interests calculated in financial sector to the money supply multiplied by some parameter. In these case that paramater is 10. I am adding it through parameter Donation_select case (left from Central bank table).
It was one of my major idea before I used the software so I wanted to simulate that. But I didn't expected that about parameter. Also, fraction reserve ratio rose from 0,1 to 0,2 which I didn't expected.
With that settings I get small inflation and Citisen 1 (agricultural) always has more money than debt, which I desired. I had to replace operator < with max function in some cases to be able to use d/dt operator.
I find that some conditions are stronger and some are weaker and tried to set conditions to be strongest about government money stock (GovermentM) and banks fractional reserve ratio (Reserves_requirement ratio). My opinion is that assumption that transactions are done imidiately is never true. All subjects will have some stock of money and that money has to be borrowed. Economical subjects to have zero stock of money is some boundary case leading to invalid conclusions.
Once I read Wynne Godley book to the end I will delete uncessary tables - balance sheet matrix and Income and expenditure matrix. They are my help tool for understanding that book.
My notes are sometimes not accurate but I left them because I see accountants as potential users of software.
Wynne Godley stressed importance of understanding of inflationary accounting but I didn't came to this point yet. Not even to pricing.
Last edit: Kresimir 2019-05-02
I guess that that parameter of value of 10 should have unit of year. First thing i thought about is 1/r i.e. 1/(interest rate). With high inflation then money added to supply would be lower and with lower inflation it would be higher. But I would further lower money added to supply with some other dimenesionless parameter (or combination of two parameter - in example velocity and something else) because I don't want hyperinflation in model.
I also used abs operator on one place without much need for it. It would make model dimensionaly not valid. It can be omitted or multiplied with number 1 of oposite unit. It want change anything.
It looks like this now with green line over red line in money and debt chart (upper left).
Inflationary accounting I could maybe add with additional row in each table with flow variables correcting the value of stocks (reevaluating) but I want do that before I get rid of balance sheet matrix and income and expenditure matrix.
I have not previous experience in modeling so I setted conditions "by eye". When value of some stock is too low or too high, subject or "sector" will do some action -in example buy bonds" in amount related to "values" of other variables from its financial statement because it will be its first information. I counted that if I connect all other variables, that will trigger other conditions. It is maybe not the best way to run model but only I could do for now. That is the reason why equtation tab is hardly readable. Parameters outside model are, in example, production and interest rates an tau variables. I will probably try to make interest rates and also tau variables changeble with inflation to see if it is true that goverment has some inflationary gain on its debt with inflation.
Lending and repaying variables I imagine as more small loans and not one loan which is payed with annuity.
As I have already written, my model is meant to better understand accounting.
Stocks of money also want be the same with inflation.
Another limitation is that citisen1 can take more loans than he need. So maybe stocks of money would have to be related to stock of production or maybe to "unused resources" or potential production. I will see if I will try to improve that. But based on this file I now understand CB financial statement, aggregated banks financial statements, sectoral financial statements and some other things better than before.
And that thing about used and unused resources and production and potential production I see as something which could explain credit crises at least roughly. Since loan creates money but payments with interests comes later.
Because "techology" or using of a "resourses" don't improve continuously but improves with "steps" and financial system has other nature.
Fractional reserve system and bank liquidity ratio and capital ratio should also be more examined and interest rates and risks banks take compared to other subjects.
Inequality is also important.
Last edit: Kresimir 2019-06-05