though your notation is questionable, I understand what you are saying. Using your definitions, If f(ET*I) < f(s), income goes up. If f(ET*I) > f(s), income goes down. That's not a difficult concept. Now here's your challenge. Define the equation that maps the domain of f(ET*I) to its range. Do the same for f(s)
I agree if it doesn't exist the significance is irrelevant because there is no significance. I agree that a 100% tax rate would likely have an effect on gross income. I've stated my position at least twice already in this thread. Empirical evidence suggests that the wide variations we've had in tax policy over the past 60 years has revealed no significant correlation between tax policy and economic growth. Therefore the logical implication is that the taxes we've had over the past 60 years have had no or no significant effect on the economy. Nor have I seen any particularly strong argument for why this would simply be a massive coincidence. Population growth has not changed so radically as to suggest there adjusting for it would demonstrate there is any strong correlation. Probably not. You could make an argument that prudent, competent fiscal management by the government creates a more positive business environment, but it is not something I could prove. Probably not. You could make an argument that imprudent, incompetent fiscal management by the government creates a more negative business environment, but it is not something I could prove.
Well, I don't think using code is the best or clearest way to describe this, but fine: double R, ET, S, P, A, PI, T, D, TE; double PET; //the previous effective tax rate int Net_Effect; //net effect of taxation and spending on national income (as either negative, positive, or neutral) int main //temporarily placed here for readability { P = <some_constant> (current population in households) A = Average Household income (base) PI = Income (P, A); T = <some_constant> (Marginal Tax Rate) D = <deductions>/PI TE = <tax_evader_losses>/PI PET = <some_constant> (previous_effective_rate) S = <some_constant> (Spending into the economy) ET = Effective_rate (T, D, TE); Net_Effect = (int)Net_Impact(S, ET, PI); if (Net_Effect == 0) //(tax_impact == spend_impact) { R = Revenue(ET, PI); cout << "Revenues will be " + R + ", and will be larger than previous revenues if ET > PET";} else if (Net_Effect > 0 && ET >= PET) //(tax_impact < spend_impact && ET >= PET) { //R = Revenue(ET, PI + <unspecified positive number>) cout << "Revenues will increase";} else if (Net_Effect < 0 && ET >= PET) //(tax_impact > spend_impact && ET >= PET) { //R = Revenue(ET, PI - <unspecified positive number>) cout << "Revenues will decrease";}//there are additional cases, but I think the above are the main ones we're concerned with so else { cout << "This is not a case I care about right now";}} double Revenue (Effective_tax_rate, Total_household_income) { double R = Effective_tax_rate * Total_household_income; return R;} double Effective_rate (T, D, TE) { double ET = T - (D + TE); return ET;} double Income (P, A) { double PI = P * A; return PI;} double Tax_Multiplier (ET, PI) { //to be defined} double Spend_Multiplier (S) { //to be defined} double Net_Impact(S, ET, PI) { double Net_Effect = Spend_Multiplier(S) - Tax_Multiplier(ET,PI); return (Net_Effect > 0) - (Net_Effect < 0); //this returns whether the impact is positive (1), negative (-1), or neutral (0) //in ideal program we would return Net_Effect and get an exact value for your CI in every case //but again, since its difficult for us to know the exact values for Spend_Multiplier(S) and Tax_Multiplier(ET,PI), this function is returning //a value representing the relationship between the outputs of the two functions instead. } /*Note that while S is a constant, S and Spend_Multiplier (S) (aka f(S)) are not necessarily the same thing, just as Tax_Multiplier (ET, PI) (f(ET*PI)) might not necessarily return ET*PI. So when you wrote "If (taxed_revenues_impact == S)" and "else if (taxed_revenues_impact < S)" those conditionals were taking spending into account but not spending's total impact on income which may be larger than the spending itself. Or, to put it another way, those two functions are spending and taxation's affect on national income, which includes, not only the initial change in income, but as well any changes in income that could be attributed to the multiplier effect. Also note that while we know that functions involving the multiplier effect will include S, ET, and PI as stated in the code, they will likely also include additional variables as well, most notably the passage of time (t) will be included as well as increased efficiency due to what was or what could have been produced using a certain amount of income. So now, hopefully you can see why spending might not be the same as spending's impact on income, and hopefully you can also see why we cannot so easily express such a relationship in terms of simple pseudo-code. But, suppose that even if we don't know the exact values that the multiplier effect produces, but that we are able still to make a determination regarding the relationship between f(S) and f(ET*PI). Would you agree that we would be able to plug in such a determination in place of the Net_Impact function's return value in the above code? So then, if you agree with that, then we have to ask ourselves how we would make that determination. Before we answer though, we need to understand why the multiplier effect affects the national income in the first place. It is because the more money people have to spend, the more they are willing and able to spend, and that spending creates income for others, who, for the same reasons, then create additional income for others. Then, it will be a safe assumption then that taking money from one area will have an equal but opposite effect of spending the same amount of money into the same part of the economy, or spending into a different part of the economy that is just as likley to spend the money on the same types of goods/services. So what we can say here, if you agree with that, is that if we tax and spend the same amount on the same people, or the same type of people, then we can substitute a 0 for the Net_Impact function return value. The other thing I believe we can say, is that if money is taxed from parts of the economy in which it is not likely to be used, and if spending into the economy of the same amount is done in places where it is very likely to be put to use, and thus increasing the propagation of a positive multiplier effect, then we can substitute in a 1 for the Net_Impact function return value, as the taxing of money that was not going to be used has no negative effects on incomes, and taxing money from areas where it is unlikely to be spent has less a negative effect on incomes, than does spending money into areas where it is more likely to be spent has a positive effect. Do you agree? Let me know if there is something I can explain further or if I made a mistake somewhere (there's a lot there so it would not exactly be a surprise to me if I left something out or put something where it doesn't belong). Also, as for subbing into the Net_Impact function return value, I'm not sure how to put that into pseudo-code, words seem much better suited for that part if you ask me, but feel free to try for yourself. */ -Meta
So you agree with them then. That's good. See above post. (look towards the bottom, around the emboldened paragraph) -Meta
Then you agree that there is likely some relationship between ET and I. I wasn't asking you about your position on our tax policy's effect on economic growth over the past 60 years. I was asking you if you believe that there is a relationship between ET and I, which you answered above. Your say-so is not good enough. You could make that argument, and I believe most folks would agree with you if you did. You could make that argument, and I believe most folks would agree with you if you did.
there is no statistically significant data to demonstrate that no correlation exists. There are a host of factors other than "tax policy". For instance, the economy cannot even remain flat without a continued massive influx of cash into the financial sector by the FED. a humorous, incredibly simplistic interpretation of an implication, with no statistical merit. Well bless your heart, you haven't "seen" any
I would not be the one to ask to correct your code, I am too scatter brained and heading into baby brain mode trying to keep some brain cells active. We can assume Big brains with super computers already have working programs of significant vastness, that is I assume how they come out and say so and so tax cut will do such and such so quickly, but you probably feel better thinking about it, and nothing is insurmountable. Anyway, whatever you learn in the process is useful for argument. Nobody can totally predict chaos but it doesn't hurt to get close. As to the Net_Impact function, which is the important one, and the words that seem to be impossible to code, it does not have to be perfect: It seems to me that if one wanted to come up with a constant, some number from 1 to 10 maybe, to use in an equation for a guesstimate of the effects of taxation, it is not impossible. Roughly you could probably look as the same economic indicators, not too many, for a couple of years before and after significant tax changes (maybe just the ones from Raygun to Bush2), making note of the kind of changes (income tax rates, and capital gains rates...), then you have something to compare with current data. You could have a program make comparisons between past data to find the closest mean sort of speak, to determine the number to use for that "multiplier effect." Or, maybe slap shades of lighter or darker color dots on the charts of the data and take them outside and ask people from a distance to assign a number from 1 to 10 to each chart; basically you are looking for a judgment of sorts as if looking at Romney and Obama's color, pardon the pun. Then after you have everything working, then you can mess with asking for current data and saving it; it is not necessary for your purposes of argument. You could like test the program seeing how close it comes to predicting known outcomes here and for other countries. In argument you can say your program is so many percent accurate for certain data ranges, and challenge them to do better, and ask, "do you have one that is better?" It might be nice to create a Rate_of_Income_Group_Change function, figure some way to look at past and current data to show the rate of change between the richer and the poorer within certain tax ranges over the years. http://www.ntu.org/tax-basics/history-of-federal-individual-1.html Look at it, it should pop out at you! It would be nice if there was something more robust like a USDA National Nutrient Database with regard to tax data and stuff with comma delimiters you could download to plug in like in the linux Nut program; PDF's and web pages are like useless and a pain the ass, unless you want to pluck the data yourself. Like between Raygun and Clinton the rate of change for certain groups might be more toward the rich than the poor. Then you return something you can use in a more enhanced Net_Impact function. That is getting fancy. Go with simple at first; mostly the data may not be out there for easy retrieval. There is no way to predict how a president or his generals are going to act in a Libya or a hurricane, or how people will react to them exactly. Just like spending could be on fixing things we should have fixed during a so-called "boom" and the real outcome is not noticed for years. Like you fix a bridge, nobody sees that if it had not been fixed they have to replace it sooner for more money wasted. We spent tax dollar burying power lines, I thought it was a smart move; the long term benefits probably do not show up easy. And for purposes of argument, unless the other side comes up with more effort and something better, in rhetorical terms, how people view your effort compared to their effort, you could win. It is not like I want Obama to win, I just want America to win. Like I said once after Obama's election it is in our best interests if he goes down as a good president. I think he failed, when he had both houses, but just like when I voted for McCain, and said he deserve to loose after three debates where he could not answer Obama's question on why invade with no WMD, if Romney fails it is still Not Camelot stupid. And if Romney wins and gets some of what he wants, you get more data to make him a one term wonder; while we scream and holler that the Democratic Congress was the Party of "NO." At some point hopefully we can come together and fix this economy, which impacts our lives the most--regardless of whether a woman's legs spreading needs a gubermint funded chastity belt or the Gays wanting one earth same laws principle guiding our big heads with the little ones.
Revenues are just a function of the individual paying (or not paying) taxes. The variables that determine individual behavior are manifold and nebulous. Functions are just constructs of math - they are not magic, they have limitations. There really are some systems that are too chaotic or subjective to be modeled with a function.
At the extremes, probably. Oh well. Since you claim population is a factor, maybe you could adjust for population data and show us whether that, for example, explains why the economy kicked ass with Clinton's higher taxes and sucked with Bush's lower taxes. I doubt it. Could be.
Sure there is. We have at least 60 years worth of data. And if you go back even further you'll find that the low rates of the late 20s (26% top marginal tax rate) also included the Great Depression, while the high rate passed in 1932 (63%) and higher rates thereafter corresponded with strong growth in the '30s and '40s Of course. And the the data suggests strongly they have far more effect than tax policy. You'd have to show me the data to believe that claim. But as you point out, that is a different factor than tax policy. Which you humorous have done absolutely nothing to rebut other than your own, baseless, conclusory say so. Correct. And I still haven't seen any particularly strong argument for why this would simply be a massive coincidence after you post. Just your own, baseless, conclusory say so.
It should not be all that hard to mix some historical matching of data, and things like the approval ratings of the congress and president with simple math. Certainly you cannot predict that the Bush Tax cuts would not work as well for Obama without looking at Business is afraid of the taxes he has coming effect. So include proposed taxes, or an "Obama factor"; might be easier to develop after the election of Romney. It does not have to be 100% accurate to be a useful Eight Ball app. Just slide up or down on tax sliders, depending upon what the candidate said, maybe have Fair Tax crap too, and a Romney rand() for deductions, and shake for a cute retort. You would think people cannot build airplanes or predict how far the boat will move if you push it...{I have totally forgot that 151 Calculus stuff}
we have a handful of tax policy changes over that time. Statistically, that amounts to nothing. I'd have to show you that there are many variables ? Very observant. I just told you it was a variable, other than tax rates, that affects revenue. That's the point of "other variables", they are different. Yet, in a multivariable system, they all interact to affect the output. You made the claim. It is up to you to demonstrate the significance of the data.
n=80 or so is plenty to make statistical deductions. Sure, I agree that many other variables affect the economy. But tax policy in the ranges we've had over the past 60 years isn't one of them, the data suggests. I have. You made the claim that tax policy affects income. It is up to you to demonstrate the significance of the data.
If the relationship likely exists at the extremes, then that means it likely exists elsewhere. Absolutely nothing in economics regarding human behavior can be described by discrete models as you are suggesting. Why would I do that? All I was doing was showing that your data and conclusion is meaningless if you don't adjust for population changes.
Tell me about the tests that you have applied to deduce significance in this complex multivariable system this is not a single variable system. You cannot isolate tax policy from the rest. Life is not the partial differential. I claimed that you cannot predict the effect. Please be honest.
None. You can always measure correlation or lack thereof. So now you are changing your story and you do not claim there is any effect on income from taxes?
None. You can always measure correlation or lack thereof. So now you are changing your story and you do not claim there is any effect on income from taxes?
So what? It can be. Lack of correlation suggests no causation. Good for you. But unless you are claiming there is an negative effect on gross income from increased taxes, then an increase in the effective tax rate will result in higher revenues for whatever level of gross income is achieved.
lots of things "can be" counselor. I just told you what I claimed counselor. You are the only one who made a predictive claim.
You have already admitted that it's likely that there is a relationship between ET and I "at the extremes". Do you believe that there is no relationship between ET and I at the non-extremes? Or do you believe that we cannot know if the relationship exists at the non-extremes? You haven't prove that the difference is insignificant enough to affect your conclusion.
