Interest Rates and Saving

Martin Feldstein says:

The Federal Reserve has reversed its low interest rate policy... It will only be a matter of time until the household saving rate is at least back to the 2.4 per cent level of 2002.

I was curious about this so I plotted the 3-month T-Bill rate against the personal saving rate since 1980 to get a rough idea of the association between the two variables:

Click to enlarge

Comments

[pgl]

In defense of Feldstein, correlations of 2 endogenous variables tells us little about the structural relationship between the 2.

[Mark Thoma]

Yeah, hence the word "rough." But I thought this was "roughly" supportive of a relationship.

[Winslow R.]

Decreasing price of gold, decreasing inflation rate, increasing home price, decreasing 'real' minimum wage etc. or my personal favorite, increasing wealth concentration is associated with the savings rate decline in 80's to 00's as well, are they causal?

Extending graph back to the 50's through 70's 'messes' up your graph but perhaps would clarify true cause?

I will read 25 page NIPA paper when I have time, but for now what exactly are these savings composed of? I assume this is savings of nonfinancial assets? As all financial assets net to zero if you include future tax liabilities as an offset to any reserves (paper dollars, tsy secs) in circulation.

Or is this a graph of savings of future tax liabilities by the U.S. private sector in competition with the foreign sector with the government sector as the producer of those liabilities through deficit spending?

Admits to the confusion: http://en.wikipedia.org/wiki/Savings_rate

To help establish whether an asset is saving(s) or an investment you shoud ask yourself where is my money invested. If the answer is cash then it is savings, if it is a type of asset which can fluctuate in nominal value then it is investment.

From this definition only short-term cash would be savings. I would extend this definition to include all types of future gov tax liabilities (tsy secs).

[reason]

I would have thought that the real interest rate was the relevant variable. How carefully did you chose you time frame Mark?

[ilsm]

I have a credit card where I get a $1.25 back for each $100 I spend.

Might as well............ toss it away than put it in the bank.

Inflation prospects..........

[pgl]

Mark - late at night I got confused as to the purported sign. You are right - savings and interest rates were moving in the same direction.

[spencer]

From 19860 to 1981 the savings rate rose and since 1981 it has fallen.

You can find many things that correlated with that but
the best seems to be the stock and housing markets.

If the best explanation for the low savings rate is the
long term bull market in assets that increased wealth
without the need to have high annual contributions -- savings -- the correlation with short term interest rates could be expained via that mechanism.

If ones long term goal is have a million dollar in stocks at retirement it is easy to calculate the
annual contribution you would need to make at various
rates of return. If you raise the rate of return
it reduces the needed annual contribution. So a stock market bull market could be the best explanation for low savings and since stock returns are generally
negatively correlated with interest rates this mechanism would explain the conclusion that savings are negatively correlated with rates.

[Robert]

Spencer said:

If the best explanation for the low savings rate is the long term bull market in assets that increased wealth without the need to have high annual contributions -- savings --the correlation with short term interest rates could be expained via that mechanism.

Is a long-term bull-market in assets the same thing as a long-term bear market in the value of money?!?!

[anne]

Please, a long term bull market in asset values is precisely a long term bull market but there is a catch. Investors have earned shockingly little of the very real bull market in stocks and bonds these last 30 years. The S&P stock index has returned 12.2% a year for 30 years after Vanguard costs, while the Vanguard long term investment-grade bond fund has returned about 8.8% a year. Investors however have generally gained far less because costs beyond long term Vanguard investing have been far far higher.

[Holly W.]

Is there really that much of a wealth effect from a bull market? You have to save some money in order to become an investor in the first place, and a negative savings rate doesn't seem to be boding well for that.

How do all these tax-free savings vehicles affect the official savings rate? If I receive a capital gain or dividend distribution from a taxable mutual fund, I have to report it as income. But I don't have to report distributions from my IRA or 401k on my 1040, so how is this money figured into the personal savings rate?

[Holly W.]

Clarification: I mean capital gain/dividend distributions that stay in my IRA or 401k, not distributions taken out of them. "From" was not really the right word there ...

[anne]

Holly's argument is clever, and I would only argue that the recent housing boom "may" be different in confidence building effect than the bull stock, bond and commercial real estate markets. Interest rates through the 1950s and 1960s were lower than now, but savings rates were higher. Also, as we have discussed savings in interest earning assets has been tax disadvantaged relative to dividends and capital gains.

[vorpal]

Could increasing interest rates cause personal saving to go further down? At least temporarily?
If you are not in debt, then higher interest rates give greater incentive to save.
But, if somebody is in debt, and the interest charges go up, then they have less money to save, because of the extra debt payments.
Just a thought.

[Holly W.]

Assuming Spencer's argument is correct and people are coasting on the belief that their current investments will provide enough income in retirement so long as the market keeps rising at X percent per year: what keeps the stock market going up if everyone thinks this way and no-one is adding any new money to it, hmmmm?

Recent news articles indicate that many Americans suddenly feel that they aren't saving enough money for retirement after all, so that trend may boost the personal savings rate no matter what future interest rates may be.

[Robert]

What is it that distinguishes a bull market in assets from a bear market in the value of money?

A long-term bear market in the value of money would show debasement against most goods, services, minor & secondary asset classes outside of shares & real estate. And this is seems to ring true (for US)outside the set of manufactured goods that have benefited from the largest of productivity gains and/or production economies of scale, or the lowest of low-cost Asian labour. PGL must have the numbers somewhere showing the "genuine" compounded inflation rate (ex quality adjustments and other artifical BLS depressors intended to limit CoLA indexation on entitlements)to compare against financial asset returns alone. This comparison might demonstrate that the bear market in the value of money is as reflective (if not more) of the bullmarket in asset prices - considering also that one would want to consider the current level of ex-ante equity risk-premium (for stocks) and the recent extended period of low nominal rates coincidental to low & negative real rates (which can be seen as the elevator of real-estate prices).

[James Kroeger]

Spencer:So a stock market bull market could be the best explanation for low savings... In support of this point I thought I'd remind everyone that because the BEA's calculation of the Personal Savings Rate does not include capital gains income, we can expect the measured personal savings rate to drop whenever capital gains income increases as a percentage of total income.

Savings = Income – Consumption

PSR = Savings / Income

PSR = (Income – Consumption) / Income

During a bull market, at least some money that would otherwise have been 'saved' in a bank is used to bid up the prices of assets instead, increasing somebody's income. Note that a household would not necessarily experience any increase in its 'consumption' when it reduces its 'savings' in this way.

This is one big reason why the Personal Savings Rate provides us little or no useful information re: the nation's savings (i.e., consumption) habits. The drop that we've seen in the Personal Savings Rate has not been the result of households 'consuming too much.' It has merely been the result of rich people changing the composition of their portfolios.

I find it really annoying whenever economists cite the Personal Saving Rate as a reason for Americans to "change their profligate ways." There is no shortage of savings in the American economy today.

[anne]

Actually, though I wish it were so, it is not so; unrealized capital gains are not making up for the low saving rate, for a continually surprising lack of saving :) The wealthy do save, but not enough to balance the lack of saving by the less wealthy. Only corporate savings are healthy, quite healthy.

[pgl]

OK, the Baby Angrybear (me) finally woke up, got his thinking cap on, and tried to post something intelligent about this. Enjoy (I hope).

[Emmanuel]

I'm with Anne on this one.

Also, note that the trend of monthly dissavings started roughly at the same time as when the Fed started its 25 BP hikes from a 1.0% Fed funds rate. Feldstein's argument is basically this: people will save more as interest earned on savings increases. Perhaps so, but I think we also need to factor in, among other things, (1) the cost of living is rising, and at the same time (2) income growth has been stagnant.

We don't have to reflect much here: a higher cost of living not offset with increases in income will tend to result in dwindling savings. Higher interest rates on savings accounts should help, but what if there's nothing left over to save in the first place?

[Robert]

Emmanuel said
but what if there's nothing left over to save in the first place?

IF this is so for those at the "wrong" end of the skewed distribution of aggregate savings, then tax-rate normalization - particularly on the upper marginal rates - as well as other targeted taxes (wealth taxes, luxury taxes) seems to be the policy of choice that would provide more harmonious fiscal balance, lower equilibrium rates, with the least macroeconomic dislocation.

But the question remains: is the "nothing left" - on average - attributable more to excessive consumption or as a result of shortfalls in income? They might appear to represent the same thing, but have markedly different philosophic and macroeconomic policy implications.